06 Probabilisitc Graphical Models Part 2: From Bayesian Networks to Graph Bandits<\/title><series>652fa592-2a86-43cb-8a9c-1c439dd00854<\/series><seriestitle>[709049] Medical Informatics<\/seriestitle><creators><creator>Holzinger A<\/creator><\/creators><license>Creative Commons 3.0: Attribution-NonCommercial-NoDerivs<\/license><media><track id=\"063565b9-8e04-432e-afdc-6f78f2b96faa\" type=\"presentation/delivery\" ref=\"track:ffd0aa9a-268d-4895-9d85-55a37afe0fa5\"><mimetype>video/mp4<\/mimetype><tags><tag>engage-download<\/tag><tag>engage-streaming<\/tag><\/tags><url>https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/7b4367f8-ded8-42dd-8367-5b5385b8021f/track.mp4<\/url><checksum type=\"md5\">d138a82adf658c4de2d7dbfcd4ce4007<\/checksum><duration>5414987<\/duration><audio id=\"audio-1\"><device/><encoder type=\"AAC (Advanced Audio Coding)\"/><channels>2<\/channels><samplingrate>44100<\/samplingrate><bitrate>64000.0<\/bitrate><\/audio><video id=\"video-1\"><device/><encoder type=\"H.264 / AVC / MPEG-4 AVC / MPEG-4 part 10\"/><bitrate>202125.0<\/bitrate><framerate>25.0<\/framerate><resolution>1280x800<\/resolution><\/video><live>false<\/live><\/track><track id=\"45210525-a5a8-4bab-a0a7-9451dfe8561e\" type=\"presentation/delivery\" ref=\"track:ffd0aa9a-268d-4895-9d85-55a37afe0fa5\" transport=\"RTMP\"><mimetype>video/mp4<\/mimetype><tags><tag>engage-download<\/tag><tag>engage-streaming<\/tag><\/tags><url>rtmp://wowza.tugraz.at/matterhorn_engage/mp4:mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/7b4367f8-ded8-42dd-8367-5b5385b8021f/track<\/url><checksum type=\"md5\">d138a82adf658c4de2d7dbfcd4ce4007<\/checksum><duration>5414987<\/duration><audio id=\"audio-1\"><device/><encoder type=\"AAC (Advanced Audio Coding)\"/><channels>2<\/channels><samplingrate>44100<\/samplingrate><bitrate>64000.0<\/bitrate><\/audio><video id=\"video-1\"><device/><encoder type=\"H.264 / AVC / MPEG-4 AVC / MPEG-4 part 10\"/><bitrate>202125.0<\/bitrate><framerate>25.0<\/framerate><resolution>1280x800<\/resolution><\/video><live>false<\/live><\/track><\/media><metadata><catalog id=\"cd56c60b-24f4-435d-adb8-92302f589a17\" type=\"dublincore/series\"><mimetype>text/xml<\/mimetype><tags><tag>archive<\/tag><tag>engage-download<\/tag><\/tags><url>https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/70a3ffa5-3225-410b-9c34-f2fa4decb6c7/70a3ffa5_3225_410b_9c34_f2fa4decb6c7.xml<\/url><checksum type=\"md5\">4caaf586bd93d89f50877dc9fa14b3dd<\/checksum><\/catalog><catalog id=\"b6c0881d-f91f-490d-b811-5014abfaf960\" type=\"dublincore/episode\"><mimetype>text/xml<\/mimetype><tags><tag>archive<\/tag><tag>engage-download<\/tag><\/tags><url>https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/4e36ced5-957c-48ea-8630-7b849984ca65/4e36ced5_957c_48ea_8630_7b849984ca65.xml<\/url><checksum type=\"md5\">5ee09835528d7136a96a56ed644c1e2a<\/checksum><\/catalog><catalog id=\"94c5ffb5-dd02-4494-b309-a71abde8ef14\" type=\"mpeg-7/text\" ref=\"track:ffd0aa9a-268d-4895-9d85-55a37afe0fa5\"><mimetype>text/xml<\/mimetype><tags><tag>engage-download<\/tag><\/tags><url>https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/9183f8ea-050f-41b3-9ba4-908e2d23a78e/slidetext.xml<\/url><\/catalog><catalog id=\"4bd8ad77-b8f5-416a-895f-18dd34257b7a\" type=\"mpeg-7/text\" ref=\"track:track-1\"><mimetype>text/xml<\/mimetype><tags><tag>archive<\/tag><tag>engage-download<\/tag><\/tags><url>https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/395f6144-4ec0-4ba4-ac5f-3df8215b07a2/catalog.xml<\/url><checksum type=\"md5\">f1597daed7f0f4d4bbd92da9c5634a16<\/checksum><\/catalog><\/metadata><attachments><attachment id=\"973e1493-1639-4a15-bbac-7b5699a858d1\" type=\"security/xacml\"><mimetype>text/xml<\/mimetype><tags><tag>archive<\/tag><\/tags><url>https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/security-policy-series/security-policy-series.xml<\/url><checksum type=\"md5\">4fe341748523f9c7035c9af3ecb78579<\/checksum><\/attachment><attachment id=\"f0b2f9e4-f4fc-4702-987d-d912e7a17f8b\" type=\"security/xacml+series\"><mimetype>text/xml<\/mimetype><tags/><url>https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/dd5b3ad7-c339-46ec-994d-40a3b9aa59d0/xacml.xml<\/url><size>0<\/size><\/attachment><attachment id=\"d6152ef5-0a06-4853-9368-85cd77d36a44\" type=\"presentation/segment+preview\" ref=\"track:45210525-a5a8-4bab-a0a7-9451dfe8561e;time=T00:21:58:560F1000\"><mimetype>image/jpeg<\/mimetype><tags><tag>engage-download<\/tag><\/tags><url>https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/1336c585-1504-49a0-a7bf-0fc9572f8c67/track.jpg<\/url><size>0<\/size><\/attachment><attachment id=\"3f16fecc-bea3-45b6-b610-213597caead5\" type=\"presentation/segment+preview\" ref=\"track:45210525-a5a8-4bab-a0a7-9451dfe8561e;time=T00:24:52:520F1000\"><mimetype>image/jpeg<\/mimetype><tags><tag>engage-download<\/tag><\/tags><url>https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/f00a235f-714a-419a-b93b-f9c733c277ee/track.jpg<\/url><size>0<\/size><\/attachment><attachment id=\"863b769a-b7a6-4904-bc51-c23ef55fdb00\" type=\"presentation/segment+preview\" ref=\"track:45210525-a5a8-4bab-a0a7-9451dfe8561e;time=T00:29:54:920F1000\"><mimetype>image/jpeg<\/mimetype><tags><tag>engage-download<\/tag><\/tags><url>https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/c579ee9b-7e66-4ac9-931d-55243d9236bc/track.jpg<\/url><size>0<\/size><\/attachment><attachment id=\"0d493e38-7153-4332-bc9d-fef87b80affa\" type=\"presentation/segment+preview\" ref=\"track:45210525-a5a8-4bab-a0a7-9451dfe8561e;time=T01:06:51:840F1000\"><mimetype>image/jpeg<\/mimetype><tags><tag>engage-download<\/tag><\/tags><url>https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/68c1787c-db47-4066-80ee-185a2205586c/track.jpg<\/url><size>0<\/size><\/attachment><attachment id=\"cb4856d6-94d6-4287-9831-c3f7de57fc42\" type=\"presentation/segment+preview\" ref=\"track:45210525-a5a8-4bab-a0a7-9451dfe8561e;time=T00:39:31:280F1000\"><mimetype>image/jpeg<\/mimetype><tags><tag>engage-download<\/tag><\/tags><url>https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/9ccda672-3ca8-40f7-a797-86c4ccc1daf6/track.jpg<\/url><size>0<\/size><\/attachment><attachment id=\"9480c549-73b0-4eb8-b108-a436230a608d\" type=\"presentation/segment+preview\" ref=\"track:45210525-a5a8-4bab-a0a7-9451dfe8561e;time=T00:49:59:120F1000\"><mimetype>image/jpeg<\/mimetype><tags><tag>engage-download<\/tag><\/tags><url>https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/c342a425-cabb-40a0-a732-535300542d22/track.jpg<\/url><size>0<\/size><\/attachment><attachment id=\"c0d5715d-1f0f-45c2-9fd6-b37faa14ed26\" type=\"presentation/segment+preview\" ref=\"track:45210525-a5a8-4bab-a0a7-9451dfe8561e;time=T00:17:14:839F1000\"><mimetype>image/jpeg<\/mimetype><tags><tag>engage-download<\/tag><\/tags><url>https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/540877d1-ab40-486f-a793-75be2b000005/track.jpg<\/url><size>0<\/size><\/attachment><attachment id=\"3ee01ed2-2025-4a76-9623-791bac9479c2\" type=\"presentation/segment+preview\" ref=\"track:45210525-a5a8-4bab-a0a7-9451dfe8561e;time=T01:28:37:360F1000\"><mimetype>image/jpeg<\/mimetype><tags><tag>engage-download<\/tag><\/tags><url>https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/71f34059-5140-44d2-bfc5-81624434cbd0/track.jpg<\/url><size>0<\/size><\/attachment><attachment id=\"fb6623e1-d80e-42c7-8cfa-43adc4e0ddc1\" type=\"presentation/segment+preview\" ref=\"track:45210525-a5a8-4bab-a0a7-9451dfe8561e;time=T00:27:43:320F1000\"><mimetype>image/jpeg<\/mimetype><tags><tag>engage-download<\/tag><\/tags><url>https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/82285407-758b-480f-ae22-6a0e64912c2a/track.jpg<\/url><size>0<\/size><\/attachment><attachment id=\"d0fc69bb-f012-4dd6-8bb0-de952e6a6290\" type=\"presentation/segment+preview\" ref=\"track:45210525-a5a8-4bab-a0a7-9451dfe8561e;time=T00:16:40:960F1000\"><mimetype>image/jpeg<\/mimetype><tags><tag>engage-download<\/tag><\/tags><url>https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/db18f079-68ae-447e-8e07-892c6773e8a2/track.jpg<\/url><size>0<\/size><\/attachment><attachment id=\"80ed5945-c89a-4382-9a3c-b30b5b41f26d\" type=\"presentation/segment+preview\" ref=\"track:45210525-a5a8-4bab-a0a7-9451dfe8561e;time=T00:14:12:360F1000\"><mimetype>image/jpeg<\/mimetype><tags><tag>engage-download<\/tag><\/tags><url>https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/4be1e541-ff90-44b7-8f84-c116cccc87b1/track.jpg<\/url><size>0<\/size><\/attachment><attachment id=\"e80c368c-2240-4ec7-8ac8-278724c59962\" type=\"presentation/segment+preview\" ref=\"track:45210525-a5a8-4bab-a0a7-9451dfe8561e;time=T01:05:08:880F1000\"><mimetype>image/jpeg<\/mimetype><tags><tag>engage-download<\/tag><\/tags><url>https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/835ca2af-7235-4911-aa10-c3442fcbf6a6/track.jpg<\/url><size>0<\/size><\/attachment><attachment id=\"25024805-77f0-437a-9045-d4fdc08181f6\" type=\"presentation/segment+preview\" ref=\"track:45210525-a5a8-4bab-a0a7-9451dfe8561e;time=T00:12:20:280F1000\"><mimetype>image/jpeg<\/mimetype><tags><tag>engage-download<\/tag><\/tags><url>https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/15d297a9-96b7-41dd-894d-6e4fa92d1a84/track.jpg<\/url><size>0<\/size><\/attachment><attachment id=\"ff43a14f-c068-41ad-ba53-5b94a4758a58\" type=\"presentation/segment+preview\" ref=\"track:45210525-a5a8-4bab-a0a7-9451dfe8561e;time=T01:07:23:680F1000\"><mimetype>image/jpeg<\/mimetype><tags><tag>engage-download<\/tag><\/tags><url>https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/1e052c0a-5870-4dbe-ad1b-3c15715ccbd2/track.jpg<\/url><size>0<\/size><\/attachment><attachment id=\"b50bdcd3-9fef-425f-ba28-d03488e5c01c\" type=\"presentation/segment+preview\" ref=\"track:45210525-a5a8-4bab-a0a7-9451dfe8561e;time=T00:43:10:920F1000\"><mimetype>image/jpeg<\/mimetype><tags><tag>engage-download<\/tag><\/tags><url>https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/e9325995-b460-4c65-9c8a-72f0e3972b25/track.jpg<\/url><size>0<\/size><\/attachment><attachment id=\"6272e0b8-5f77-43c2-a2ae-85c64813deff\" type=\"presentation/segment+preview\" ref=\"track:45210525-a5a8-4bab-a0a7-9451dfe8561e;time=T01:26:33:920F1000\"><mimetype>image/jpeg<\/mimetype><tags><tag>engage-download<\/tag><\/tags><url>https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/e3960148-e0b0-45c3-84ac-b8fc8bcd4ea1/track.jpg<\/url><size>0<\/size><\/attachment><attachment id=\"b0883db4-e094-4689-b42c-9a3e6ddd4d43\" type=\"presentation/segment+preview\" ref=\"track:45210525-a5a8-4bab-a0a7-9451dfe8561e;time=T00:20:56:40F1000\"><mimetype>image/jpeg<\/mimetype><tags><tag>engage-download<\/tag><\/tags><url>https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/a4fd407c-ab7b-4a5e-8434-7dcb1100fd5e/track.jpg<\/url><size>0<\/size><\/attachment><attachment id=\"f6ddd8d0-9557-421f-8fa2-0a11d3d51eab\" type=\"presentation/segment+preview\" ref=\"track:45210525-a5a8-4bab-a0a7-9451dfe8561e;time=T01:24:53:80F1000\"><mimetype>image/jpeg<\/mimetype><tags><tag>engage-download<\/tag><\/tags><url>https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/e3f0f740-6fc8-415b-bba3-6a248bc21de2/track.jpg<\/url><size>0<\/size><\/attachment><attachment id=\"36d183af-6fe9-4a69-9d51-e43dd53325f2\" type=\"presentation/segment+preview\" ref=\"track:45210525-a5a8-4bab-a0a7-9451dfe8561e;time=T00:52:04:880F1000\"><mimetype>image/jpeg<\/mimetype><tags><tag>engage-download<\/tag><\/tags><url>https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/d6759621-7816-423d-9969-dd4cccb1b4ba/track.jpg<\/url><size>0<\/size><\/attachment><attachment id=\"4ed36e18-5feb-495c-8c33-7e2ebe094952\" type=\"presentation/segment+preview\" ref=\"track:45210525-a5a8-4bab-a0a7-9451dfe8561e;time=T00:12:13:800F1000\"><mimetype>image/jpeg<\/mimetype><tags><tag>engage-download<\/tag><\/tags><url>https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/15e202d1-27b0-496e-ac89-ca801fca8e2f/track.jpg<\/url><size>0<\/size><\/attachment><attachment id=\"fee30829-93c6-4174-ab7e-aeb51218eec1\" type=\"presentation/segment+preview\" ref=\"track:45210525-a5a8-4bab-a0a7-9451dfe8561e;time=T00:24:37:760F1000\"><mimetype>image/jpeg<\/mimetype><tags><tag>engage-download<\/tag><\/tags><url>https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/6caf36e0-10ef-496a-adc7-2c4bc89ef2bd/track.jpg<\/url><size>0<\/size><\/attachment><attachment id=\"3ce34bd0-a5c1-4765-9141-083f32b5ba4d\" type=\"presentation/segment+preview\" ref=\"track:45210525-a5a8-4bab-a0a7-9451dfe8561e;time=T01:02:37:960F1000\"><mimetype>image/jpeg<\/mimetype><tags><tag>engage-download<\/tag><\/tags><url>https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/17e75605-f22c-49d6-9fbc-c0f4900bcfbc/track.jpg<\/url><size>0<\/size><\/attachment><attachment id=\"c7da78b9-63ef-47b4-ba14-05ae686ad8df\" type=\"presentation/segment+preview\" ref=\"track:45210525-a5a8-4bab-a0a7-9451dfe8561e;time=T01:04:28:600F1000\"><mimetype>image/jpeg<\/mimetype><tags><tag>engage-download<\/tag><\/tags><url>https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/f0de732e-5ddd-4caf-ab64-ba2e7939cb70/track.jpg<\/url><size>0<\/size><\/attachment><attachment id=\"a78a5636-c0d8-48fc-8890-d30851e28933\" type=\"presentation/segment+preview\" ref=\"track:45210525-a5a8-4bab-a0a7-9451dfe8561e;time=T01:25:52:560F1000\"><mimetype>image/jpeg<\/mimetype><tags><tag>engage-download<\/tag><\/tags><url>https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/bf4d1411-16c2-48b3-b9eb-4e720366858f/track.jpg<\/url><size>0<\/size><\/attachment><attachment id=\"7de914dd-543d-4840-9295-054a79829d9a\" type=\"presentation/segment+preview\" ref=\"track:45210525-a5a8-4bab-a0a7-9451dfe8561e;time=T00:28:40:360F1000\"><mimetype>image/jpeg<\/mimetype><tags><tag>engage-download<\/tag><\/tags><url>https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/6446e2b3-f7a8-4e64-9df1-46d63281b45b/track.jpg<\/url><size>0<\/size><\/attachment><attachment id=\"50f9aad9-c53e-4095-844a-f82781db8d58\" type=\"presentation/segment+preview\" ref=\"track:45210525-a5a8-4bab-a0a7-9451dfe8561e;time=T01:14:00:600F1000\"><mimetype>image/jpeg<\/mimetype><tags><tag>engage-download<\/tag><\/tags><url>https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/e2acd983-4a95-424d-8ad9-73111a1d8178/track.jpg<\/url><size>0<\/size><\/attachment><attachment id=\"7defadb9-2ca1-48b0-883e-57995fa247fa\" type=\"presentation/segment+preview\" ref=\"track:45210525-a5a8-4bab-a0a7-9451dfe8561e;time=T00:32:12:280F1000\"><mimetype>image/jpeg<\/mimetype><tags><tag>engage-download<\/tag><\/tags><url>https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/b29a2412-d7de-4f97-8434-0025dcb4a7a3/track.jpg<\/url><size>0<\/size><\/attachment><attachment id=\"ad14abad-e5cc-4d90-8fdc-fc46968ebaf7\" type=\"presentation/segment+preview\" ref=\"track:45210525-a5a8-4bab-a0a7-9451dfe8561e;time=T01:20:21:40F1000\"><mimetype>image/jpeg<\/mimetype><tags><tag>engage-download<\/tag><\/tags><url>https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/98f14e9e-0bd5-4814-accc-27dd0c253161/track.jpg<\/url><size>0<\/size><\/attachment><attachment id=\"097e20ed-aa55-4a8b-a26e-c3c7db3f1921\" type=\"presentation/segment+preview\" ref=\"track:45210525-a5a8-4bab-a0a7-9451dfe8561e;time=T00:45:48:400F1000\"><mimetype>image/jpeg<\/mimetype><tags><tag>engage-download<\/tag><\/tags><url>https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/6ec9c41f-f1dd-41d8-966c-cc852d005075/track.jpg<\/url><size>0<\/size><\/attachment><attachment id=\"3c86b5ca-4260-4b52-a07a-864bd48965b0\" type=\"presentation/segment+preview\" ref=\"track:45210525-a5a8-4bab-a0a7-9451dfe8561e;time=T00:01:54:560F1000\"><mimetype>image/jpeg<\/mimetype><tags><tag>engage-download<\/tag><\/tags><url>https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/274b400c-9e92-4b98-8b8a-52c3b852ad03/track.jpg<\/url><size>0<\/size><\/attachment><attachment id=\"3b56ec9d-62fa-46ec-abe5-70becc1bbe92\" type=\"presentation/segment+preview\" ref=\"track:45210525-a5a8-4bab-a0a7-9451dfe8561e;time=T00:08:17:360F1000\"><mimetype>image/jpeg<\/mimetype><tags><tag>engage-download<\/tag><\/tags><url>https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/6e951bc3-5e85-4aa8-a0b2-5238edd24b13/track.jpg<\/url><size>0<\/size><\/attachment><attachment id=\"5a477fec-238c-4117-82f0-18ce3dba34d4\" type=\"presentation/segment+preview\" ref=\"track:45210525-a5a8-4bab-a0a7-9451dfe8561e;time=T01:03:05:600F1000\"><mimetype>image/jpeg<\/mimetype><tags><tag>engage-download<\/tag><\/tags><url>https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/82b5188e-6bfa-463f-a815-e6bfc091c6eb/track.jpg<\/url><size>0<\/size><\/attachment><attachment id=\"f8ec2f9f-3bba-4f61-b56f-fb740a7427cb\" type=\"presentation/segment+preview\" ref=\"track:45210525-a5a8-4bab-a0a7-9451dfe8561e;time=T00:09:43:480F1000\"><mimetype>image/jpeg<\/mimetype><tags><tag>engage-download<\/tag><\/tags><url>https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/e8c5b1bf-d1de-4806-b6ac-a26482b5deac/track.jpg<\/url><size>0<\/size><\/attachment><attachment id=\"db089d92-a64b-4dd0-9b97-42bc5d8bc434\" type=\"presentation/segment+preview\" ref=\"track:45210525-a5a8-4bab-a0a7-9451dfe8561e;time=T00:33:59:720F1000\"><mimetype>image/jpeg<\/mimetype><tags><tag>engage-download<\/tag><\/tags><url>https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/39335258-3766-42e2-b74b-f21bfc4b5a12/track.jpg<\/url><size>0<\/size><\/attachment><attachment id=\"8689b77a-2344-461e-8032-d6d3cc58a2df\" type=\"presentation/segment+preview\" ref=\"track:45210525-a5a8-4bab-a0a7-9451dfe8561e;time=T00:34:36:40F1000\"><mimetype>image/jpeg<\/mimetype><tags><tag>engage-download<\/tag><\/tags><url>https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/be112ef5-6009-4bc3-a476-4ecb3a9d77a0/track.jpg<\/url><size>0<\/size><\/attachment><attachment id=\"4e9e957f-0d56-4861-bde9-2dc207ed2fdf\" type=\"presentation/segment+preview\" ref=\"track:45210525-a5a8-4bab-a0a7-9451dfe8561e;time=T00:19:19:480F1000\"><mimetype>image/jpeg<\/mimetype><tags><tag>engage-download<\/tag><\/tags><url>https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/a80a39c5-c445-4304-882f-6887423f6e96/track.jpg<\/url><size>0<\/size><\/attachment><attachment id=\"4f7d419a-1af0-4d24-9af2-b6e528c8facc\" type=\"presentation/segment+preview\" ref=\"track:45210525-a5a8-4bab-a0a7-9451dfe8561e;time=T00:09:16:200F1000\"><mimetype>image/jpeg<\/mimetype><tags><tag>engage-download<\/tag><\/tags><url>https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/3415228d-5667-4ed7-8379-69d95befa033/track.jpg<\/url><size>0<\/size><\/attachment><attachment id=\"15fc84d6-d6ad-4e0f-aa06-33c79a9ce129\" type=\"presentation/segment+preview\" ref=\"track:45210525-a5a8-4bab-a0a7-9451dfe8561e;time=T01:04:55:280F1000\"><mimetype>image/jpeg<\/mimetype><tags><tag>engage-download<\/tag><\/tags><url>https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/ca1ad350-3192-41f2-b2be-734b1bba7373/track.jpg<\/url><size>0<\/size><\/attachment><attachment id=\"b1ba5ad4-88fb-4ec1-82eb-3f74e5d9b39f\" type=\"presentation/segment+preview\" ref=\"track:45210525-a5a8-4bab-a0a7-9451dfe8561e;time=T01:24:59:640F1000\"><mimetype>image/jpeg<\/mimetype><tags><tag>engage-download<\/tag><\/tags><url>https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/7020a465-0d40-4eb6-8454-6ecbcd819a40/track.jpg<\/url><size>0<\/size><\/attachment><attachment id=\"bb5a0700-1b5b-4bdb-b06a-842ea9355d10\" type=\"presentation/segment+preview\" ref=\"track:45210525-a5a8-4bab-a0a7-9451dfe8561e;time=T00:49:03:680F1000\"><mimetype>image/jpeg<\/mimetype><tags><tag>engage-download<\/tag><\/tags><url>https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/6dfc7169-200b-4084-b780-bfd15c6e3651/track.jpg<\/url><size>0<\/size><\/attachment><attachment id=\"3f235ee8-fd35-4678-ad0c-90faaeb3b88a\" type=\"presentation/segment+preview\" ref=\"track:45210525-a5a8-4bab-a0a7-9451dfe8561e;time=T01:28:07:200F1000\"><mimetype>image/jpeg<\/mimetype><tags><tag>engage-download<\/tag><\/tags><url>https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/a1838e06-c47e-42bc-a268-38734a107ef3/track.jpg<\/url><size>0<\/size><\/attachment><attachment id=\"cc4ede90-3c22-4a0b-ba64-129cb94c97e7\" type=\"presentation/segment+preview\" ref=\"track:45210525-a5a8-4bab-a0a7-9451dfe8561e;time=T00:56:34:40F1000\"><mimetype>image/jpeg<\/mimetype><tags><tag>engage-download<\/tag><\/tags><url>https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/eee918ba-be51-4bd6-97a0-ab7f740bd4c7/track.jpg<\/url><size>0<\/size><\/attachment><attachment id=\"b045c314-1484-457d-bdf9-db372e84e80a\" type=\"presentation/segment+preview\" ref=\"track:45210525-a5a8-4bab-a0a7-9451dfe8561e;time=T00:00:00:0F1000\"><mimetype>image/jpeg<\/mimetype><tags><tag>engage-download<\/tag><\/tags><url>https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/d2128d70-db12-49ec-99e4-3f00f21cc51f/track.jpg<\/url><size>0<\/size><\/attachment><attachment id=\"f54e9273-e13d-45e7-98e3-babbc856ebc5\" type=\"presentation/segment+preview\" ref=\"track:45210525-a5a8-4bab-a0a7-9451dfe8561e;time=T00:48:07:520F1000\"><mimetype>image/jpeg<\/mimetype><tags><tag>engage-download<\/tag><\/tags><url>https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/dffc5f4a-3b7d-4af9-9753-04151aaca684/track.jpg<\/url><size>0<\/size><\/attachment><attachment id=\"efcc30fd-9f83-43cd-8ffe-d72462ce8faa\" type=\"presentation/segment+preview\" ref=\"track:45210525-a5a8-4bab-a0a7-9451dfe8561e;time=T00:09:32:920F1000\"><mimetype>image/jpeg<\/mimetype><tags><tag>engage-download<\/tag><\/tags><url>https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/5891df7b-cac0-4cf9-b64f-d1049234811a/track.jpg<\/url><size>0<\/size><\/attachment><attachment id=\"f3c428be-b543-4e25-a4b5-7edd15831998\" type=\"presentation/segment+preview\" ref=\"track:45210525-a5a8-4bab-a0a7-9451dfe8561e;time=T00:53:32:200F1000\"><mimetype>image/jpeg<\/mimetype><tags><tag>engage-download<\/tag><\/tags><url>https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/8c57954f-4451-45f0-98ba-131bb2f55272/track.jpg<\/url><size>0<\/size><\/attachment><attachment id=\"8aba5bc7-3ed6-4e84-aee1-7608e1dde532\" type=\"presentation/segment+preview\" ref=\"track:45210525-a5a8-4bab-a0a7-9451dfe8561e;time=T01:24:28:880F1000\"><mimetype>image/jpeg<\/mimetype><tags><tag>engage-download<\/tag><\/tags><url>https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/628b2a11-514b-4fa3-8ba9-9aca78ad4e2e/track.jpg<\/url><size>0<\/size><\/attachment><attachment id=\"7538d46d-7538-4ecf-ac09-1ab6ecb1d9e6\" type=\"presenter/player+preview\"><mimetype>image/png<\/mimetype><tags><tag>archive<\/tag><tag>engage-download<\/tag><\/tags><url>https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/e7e94061-3cfb-452e-a1c4-e6a24dd52fc5/coverimage.png<\/url><size>0<\/size><\/attachment><attachment id=\"bce001ac-79e8-449d-aa6a-b1e1785df3e1\" type=\"presentation/segment+preview\" ref=\"track:45210525-a5a8-4bab-a0a7-9451dfe8561e;time=T01:05:48:880F1000\"><mimetype>image/jpeg<\/mimetype><tags><tag>engage-download<\/tag><\/tags><url>https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/9c3c7f8a-c597-4d07-bc3b-62bae8b79223/track.jpg<\/url><size>0<\/size><\/attachment><attachment id=\"44aad8d9-6ee0-4422-a6c3-070ea404ce0d\" type=\"presentation/segment+preview\" ref=\"track:45210525-a5a8-4bab-a0a7-9451dfe8561e;time=T01:18:14:240F1000\"><mimetype>image/jpeg<\/mimetype><tags><tag>engage-download<\/tag><\/tags><url>https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/3016dfa2-6f63-4838-905a-56503c1ddea3/track.jpg<\/url><size>0<\/size><\/attachment><attachment id=\"818531d9-0657-401d-a8e0-921fa01d5639\" type=\"presentation/segment+preview\" ref=\"track:45210525-a5a8-4bab-a0a7-9451dfe8561e;time=T01:21:19:240F1000\"><mimetype>image/jpeg<\/mimetype><tags><tag>engage-download<\/tag><\/tags><url>https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/2d278a9a-1a23-449f-8cdd-718ef36b6bf5/track.jpg<\/url><size>0<\/size><\/attachment><attachment id=\"f7652451-f6fd-4a68-8d00-dada76109346\" type=\"presentation/segment+preview\" ref=\"track:45210525-a5a8-4bab-a0a7-9451dfe8561e;time=T01:12:05:0F1000\"><mimetype>image/jpeg<\/mimetype><tags><tag>engage-download<\/tag><\/tags><url>https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/d7ec3d3e-0242-4bb6-ae8e-10e8cc9d8b47/track.jpg<\/url><size>0<\/size><\/attachment><attachment id=\"6983e8ae-51b4-41e0-981a-9e6ea7ab5179\" type=\"presentation/segment+preview\" ref=\"track:45210525-a5a8-4bab-a0a7-9451dfe8561e;time=T01:01:26:560F1000\"><mimetype>image/jpeg<\/mimetype><tags><tag>engage-download<\/tag><\/tags><url>https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/661b3521-6399-4131-b8d2-bfd5c1d47441/track.jpg<\/url><size>0<\/size><\/attachment><attachment id=\"6027a2c7-db2e-4005-9645-a8a5e65cd9f6\" type=\"presentation/segment+preview\" ref=\"track:45210525-a5a8-4bab-a0a7-9451dfe8561e;time=T01:13:20:400F1000\"><mimetype>image/jpeg<\/mimetype><tags><tag>engage-download<\/tag><\/tags><url>https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/63a39edf-fc5e-43b7-8b22-dad80bdda431/track.jpg<\/url><size>0<\/size><\/attachment><attachment id=\"12f87a68-a57b-4949-a25a-4640fc043669\" type=\"presentation/segment+preview\" ref=\"track:45210525-a5a8-4bab-a0a7-9451dfe8561e;time=T00:20:14:240F1000\"><mimetype>image/jpeg<\/mimetype><tags><tag>engage-download<\/tag><\/tags><url>https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/f22280d2-d58d-4f7b-a45d-b93526d70fbe/track.jpg<\/url><size>0<\/size><\/attachment><attachment id=\"5add9872-dd04-433b-ba69-84506a3bc438\" type=\"presentation/segment+preview\" ref=\"track:45210525-a5a8-4bab-a0a7-9451dfe8561e;time=T00:54:42:680F1000\"><mimetype>image/jpeg<\/mimetype><tags><tag>engage-download<\/tag><\/tags><url>https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/5f67c6bd-1941-4052-a696-96aa17daefbc/track.jpg<\/url><size>0<\/size><\/attachment><attachment id=\"5ef67b12-360e-41d3-955e-2531c918e187\" type=\"presentation/segment+preview\" ref=\"track:45210525-a5a8-4bab-a0a7-9451dfe8561e;time=T00:54:51:960F1000\"><mimetype>image/jpeg<\/mimetype><tags><tag>engage-download<\/tag><\/tags><url>https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/35a92b3f-2e1e-4308-9c90-0a2b39f6bb51/track.jpg<\/url><size>0<\/size><\/attachment><attachment id=\"72b38d5e-8f37-4319-91fc-5807e3662ab4\" type=\"presentation/segment+preview\" ref=\"track:45210525-a5a8-4bab-a0a7-9451dfe8561e;time=T00:38:53:920F1000\"><mimetype>image/jpeg<\/mimetype><tags><tag>engage-download<\/tag><\/tags><url>https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/9af7c54f-3da7-49c3-852e-bf3b66f6aded/track.jpg<\/url><size>0<\/size><\/attachment><attachment id=\"97a40235-0821-4232-8643-655c868636d0\" type=\"presentation/segment+preview\" ref=\"track:45210525-a5a8-4bab-a0a7-9451dfe8561e;time=T00:41:14:320F1000\"><mimetype>image/jpeg<\/mimetype><tags><tag>engage-download<\/tag><\/tags><url>https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/a0c59d71-959f-4c49-ab51-5e462bf4c221/track.jpg<\/url><size>0<\/size><\/attachment><attachment id=\"92744892-71fc-4553-acc5-944b2411df4e\" type=\"presentation/segment+preview\" ref=\"track:45210525-a5a8-4bab-a0a7-9451dfe8561e;time=T01:06:09:360F1000\"><mimetype>image/jpeg<\/mimetype><tags><tag>engage-download<\/tag><\/tags><url>https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/33c4b33a-8a0a-45f8-9086-354b82a36b26/track.jpg<\/url><size>0<\/size><\/attachment><attachment id=\"04da209b-bf32-4bb9-b1e6-410eed901c4f\" type=\"presentation/segment+preview\" ref=\"track:45210525-a5a8-4bab-a0a7-9451dfe8561e;time=T01:27:30:560F1000\"><mimetype>image/jpeg<\/mimetype><tags><tag>engage-download<\/tag><\/tags><url>https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/7a161758-64bf-4e3e-97f9-825009822213/track.jpg<\/url><size>0<\/size><\/attachment><attachment id=\"e2de28e8-9bac-4c38-9ede-e20c4843dbf2\" type=\"presentation/segment+preview\" ref=\"track:45210525-a5a8-4bab-a0a7-9451dfe8561e;time=T00:42:31:360F1000\"><mimetype>image/jpeg<\/mimetype><tags><tag>engage-download<\/tag><\/tags><url>https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/f0952a7a-d650-4656-90f4-bb7e7dc5e949/track.jpg<\/url><size>0<\/size><\/attachment><attachment id=\"d05d8cf6-e4d7-4374-b1b3-2651cde1d98e\" type=\"presentation/segment+preview\" ref=\"track:45210525-a5a8-4bab-a0a7-9451dfe8561e;time=T01:09:20:799F1000\"><mimetype>image/jpeg<\/mimetype><tags><tag>engage-download<\/tag><\/tags><url>https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/4cef5745-e19a-4571-9b11-84b3948d8d8e/track.jpg<\/url><size>0<\/size><\/attachment><attachment id=\"b84c1dab-06fc-4f11-8c85-3335fc042429\" type=\"presentation/segment+preview\" ref=\"track:45210525-a5a8-4bab-a0a7-9451dfe8561e;time=T00:33:10:880F1000\"><mimetype>image/jpeg<\/mimetype><tags><tag>engage-download<\/tag><\/tags><url>https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/5b258709-ba4f-46e0-aae6-49a61a64e25b/track.jpg<\/url><size>0<\/size><\/attachment><attachment id=\"d841ccdc-4336-405a-ac19-0a381c327709\" type=\"presentation/segment+preview\" ref=\"track:45210525-a5a8-4bab-a0a7-9451dfe8561e;time=T00:26:44:960F1000\"><mimetype>image/jpeg<\/mimetype><tags><tag>engage-download<\/tag><\/tags><url>https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/519507a6-eec8-4782-b57d-9d09f351975e/track.jpg<\/url><size>0<\/size><\/attachment><attachment id=\"19acf831-1ad7-40d4-ab04-b5caa5931973\" type=\"presentation/segment+preview\" ref=\"track:45210525-a5a8-4bab-a0a7-9451dfe8561e;time=T00:31:00:480F1000\"><mimetype>image/jpeg<\/mimetype><tags><tag>engage-download<\/tag><\/tags><url>https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/83fb9e15-4f84-48c0-8ef9-0ac99e2300cc/track.jpg<\/url><size>0<\/size><\/attachment><attachment id=\"ff5278dd-824c-4e92-beaf-7e57ffcc5c90\" type=\"presentation/segment+preview\" ref=\"track:45210525-a5a8-4bab-a0a7-9451dfe8561e;time=T01:25:09:360F1000\"><mimetype>image/jpeg<\/mimetype><tags><tag>engage-download<\/tag><\/tags><url>https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/2f98a0d2-63ce-4bd0-bed8-56286b2bc93c/track.jpg<\/url><size>0<\/size><\/attachment><attachment id=\"f608aa9b-73ca-4986-8d14-b970b4696f44\" type=\"presentation/segment+preview\" ref=\"track:45210525-a5a8-4bab-a0a7-9451dfe8561e;time=T00:51:45:480F1000\"><mimetype>image/jpeg<\/mimetype><tags><tag>engage-download<\/tag><\/tags><url>https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/53f08c19-ebc5-4ce7-9ddb-2681eb42c00c/track.jpg<\/url><size>0<\/size><\/attachment><attachment id=\"d3745f34-1d3f-4738-914a-68ef9f1b6901\" type=\"presentation/segment+preview\" ref=\"track:45210525-a5a8-4bab-a0a7-9451dfe8561e;time=T01:04:37:400F1000\"><mimetype>image/jpeg<\/mimetype><tags><tag>engage-download<\/tag><\/tags><url>https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/addb0b35-fee3-4ef9-9680-e9d913c71982/track.jpg<\/url><size>0<\/size><\/attachment><attachment id=\"3480a20a-c8c4-4eee-9a63-73a75a2faa73\" type=\"presentation/segment+preview\" ref=\"track:45210525-a5a8-4bab-a0a7-9451dfe8561e;time=T01:27:38:960F1000\"><mimetype>image/jpeg<\/mimetype><tags><tag>engage-download<\/tag><\/tags><url>https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/8451a69f-11ad-4902-a25a-d412b52cf810/track.jpg<\/url><size>0<\/size><\/attachment><attachment id=\"d4b2a422-ffca-4690-80d6-84dbd79e33fe\" type=\"presentation/segment+preview\" ref=\"track:45210525-a5a8-4bab-a0a7-9451dfe8561e;time=T01:29:58:640F1000\"><mimetype>image/jpeg<\/mimetype><tags><tag>engage-download<\/tag><\/tags><url>https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/8e98bc46-c08d-4e0a-a37b-f8b7e96199f1/track.jpg<\/url><size>0<\/size><\/attachment><attachment id=\"346c06cc-569b-47e3-b84d-b8b93aa8ab38\" type=\"presentation/segment+preview\" ref=\"track:45210525-a5a8-4bab-a0a7-9451dfe8561e;time=T00:46:28:320F1000\"><mimetype>image/jpeg<\/mimetype><tags><tag>engage-download<\/tag><\/tags><url>https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/ffac8015-7952-4061-a9c5-00c95b5de942/track.jpg<\/url><size>0<\/size><\/attachment><attachment id=\"5abe3c0c-6a1b-4ab0-8818-f6866bd82ab8\" type=\"presentation/segment+preview\" ref=\"track:45210525-a5a8-4bab-a0a7-9451dfe8561e;time=T00:43:16:200F1000\"><mimetype>image/jpeg<\/mimetype><tags><tag>engage-download<\/tag><\/tags><url>https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/e62c918e-485c-45c6-9e7f-742c7e1446c4/track.jpg<\/url><size>0<\/size><\/attachment><attachment id=\"13bdb656-6534-41bd-8528-4aefce7646aa\" type=\"presentation/segment+preview\" ref=\"track:45210525-a5a8-4bab-a0a7-9451dfe8561e;time=T01:08:07:920F1000\"><mimetype>image/jpeg<\/mimetype><tags><tag>engage-download<\/tag><\/tags><url>https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/f16d5787-f7b0-4635-8675-8ea5c56411c1/track.jpg<\/url><size>0<\/size><\/attachment><attachment id=\"714ded71-ac02-4b26-9369-05670137a278\" type=\"presentation/segment+preview\" ref=\"track:45210525-a5a8-4bab-a0a7-9451dfe8561e;time=T00:26:12:0F1000\"><mimetype>image/jpeg<\/mimetype><tags><tag>engage-download<\/tag><\/tags><url>https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/e376250c-5db2-4426-b393-bbc4e7a2d19a/track.jpg<\/url><size>0<\/size><\/attachment><attachment id=\"240efa5e-dbb8-4fcf-a933-8306e345a8a2\" type=\"presentation/segment+preview\" ref=\"track:45210525-a5a8-4bab-a0a7-9451dfe8561e;time=T00:05:53:160F1000\"><mimetype>image/jpeg<\/mimetype><tags><tag>engage-download<\/tag><\/tags><url>https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/d5e51fdc-4d7c-42ce-a3de-dae9ffdb8103/track.jpg<\/url><size>0<\/size><\/attachment><attachment id=\"db521ab6-3d3c-4980-8030-e155b7afb3a7\" type=\"presentation/segment+preview\" ref=\"track:45210525-a5a8-4bab-a0a7-9451dfe8561e;time=T00:23:42:160F1000\"><mimetype>image/jpeg<\/mimetype><tags><tag>engage-download<\/tag><\/tags><url>https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/a2295e0a-f7ef-41fd-847b-40679ae28d5d/track.jpg<\/url><size>0<\/size><\/attachment><attachment id=\"d1b9dbb3-b8bf-4772-a9c2-44367f60bb15\" type=\"presentation/segment+preview\" ref=\"track:45210525-a5a8-4bab-a0a7-9451dfe8561e;time=T00:35:52:0F1000\"><mimetype>image/jpeg<\/mimetype><tags><tag>engage-download<\/tag><\/tags><url>https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/f9ba87ff-9ee0-4adc-a383-1b00646b7527/track.jpg<\/url><size>0<\/size><\/attachment><attachment id=\"634d7de0-430a-497c-99ee-239a4aa3ff53\" type=\"presentation/segment+preview\" ref=\"track:45210525-a5a8-4bab-a0a7-9451dfe8561e;time=T00:36:44:280F1000\"><mimetype>image/jpeg<\/mimetype><tags><tag>engage-download<\/tag><\/tags><url>https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/91059181-d401-4617-b548-81e0723871da/track.jpg<\/url><size>0<\/size><\/attachment><attachment id=\"ba5ef2e4-86c6-4d27-9c76-627b329265f5\" type=\"presentation/segment+preview\" ref=\"track:45210525-a5a8-4bab-a0a7-9451dfe8561e;time=T01:08:43:120F1000\"><mimetype>image/jpeg<\/mimetype><tags><tag>engage-download<\/tag><\/tags><url>https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/13803432-0117-4425-9cca-21ef610c3271/track.jpg<\/url><size>0<\/size><\/attachment><attachment id=\"8df8e840-81ff-49a5-abb2-e2bd5a3f28c0\" type=\"presentation/segment+preview\" ref=\"track:45210525-a5a8-4bab-a0a7-9451dfe8561e;time=T01:07:15:320F1000\"><mimetype>image/jpeg<\/mimetype><tags><tag>engage-download<\/tag><\/tags><url>https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/9abd0d03-0891-458b-8006-0a911695502d/track.jpg<\/url><size>0<\/size><\/attachment><attachment id=\"7e35f66c-e6ff-4936-8be3-92c4b1e9de42\" type=\"presentation/segment+preview\" ref=\"track:45210525-a5a8-4bab-a0a7-9451dfe8561e;time=T00:35:08:320F1000\"><mimetype>image/jpeg<\/mimetype><tags><tag>engage-download<\/tag><\/tags><url>https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/f1acca10-7339-48f3-b8a8-45d235c6b0c9/track.jpg<\/url><size>0<\/size><\/attachment><attachment id=\"69ed39a7-86d8-40cf-9ae8-03292ae5a785\" type=\"presentation/segment+preview\" ref=\"track:45210525-a5a8-4bab-a0a7-9451dfe8561e;time=T00:57:31:920F1000\"><mimetype>image/jpeg<\/mimetype><tags><tag>engage-download<\/tag><\/tags><url>https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/57a64a0d-b0dc-4640-8237-a4df5ef3f941/track.jpg<\/url><size>0<\/size><\/attachment><attachment id=\"f0723b44-bc94-4997-b788-9150169dca73\" type=\"presentation/segment+preview\" ref=\"track:45210525-a5a8-4bab-a0a7-9451dfe8561e;time=T00:59:43:880F1000\"><mimetype>image/jpeg<\/mimetype><tags><tag>engage-download<\/tag><\/tags><url>https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/de1e857b-3297-4b8e-8197-4734ae81cf41/track.jpg<\/url><size>0<\/size><\/attachment><attachment id=\"55c49d99-e3df-40ca-a131-fb5df173440d\" type=\"presentation/segment+preview\" ref=\"track:45210525-a5a8-4bab-a0a7-9451dfe8561e;time=T01:28:31:560F1000\"><mimetype>image/jpeg<\/mimetype><tags><tag>engage-download<\/tag><\/tags><url>https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/a521675a-8f35-40a8-acd3-8fe6991f0a3f/track.jpg<\/url><size>0<\/size><\/attachment><attachment id=\"469ec150-b26c-4ca1-a151-af9d7cea92e1\" type=\"presentation/segment+preview\" ref=\"track:45210525-a5a8-4bab-a0a7-9451dfe8561e;time=T00:35:00:960F1000\"><mimetype>image/jpeg<\/mimetype><tags><tag>engage-download<\/tag><\/tags><url>https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/ea79344c-865a-4898-970a-5f1caead5f7a/track.jpg<\/url><size>0<\/size><\/attachment><attachment id=\"52cdd696-38e5-44c4-9d7a-e4f3a69738a3\" type=\"presentation/segment+preview\" ref=\"track:45210525-a5a8-4bab-a0a7-9451dfe8561e;time=T00:25:20:520F1000\"><mimetype>image/jpeg<\/mimetype><tags><tag>engage-download<\/tag><\/tags><url>https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/6d5a6667-c1ad-476b-9d18-56ec9de25081/track.jpg<\/url><size>0<\/size><\/attachment><attachment id=\"87dbf3ff-624a-443e-8af5-417551f26a51\" type=\"presentation/segment+preview\" ref=\"track:45210525-a5a8-4bab-a0a7-9451dfe8561e;time=T00:19:25:880F1000\"><mimetype>image/jpeg<\/mimetype><tags><tag>engage-download<\/tag><\/tags><url>https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/ef3d6ebe-737c-42a6-aedd-9fc8ee781d11/track.jpg<\/url><size>0<\/size><\/attachment><attachment id=\"a08de618-f4c2-4db4-8fb8-33456baf6298\" type=\"presentation/segment+preview\" ref=\"track:45210525-a5a8-4bab-a0a7-9451dfe8561e;time=T01:24:05:480F1000\"><mimetype>image/jpeg<\/mimetype><tags><tag>engage-download<\/tag><\/tags><url>https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/1b3fc3a1-9d79-42f4-9d9c-9809b2121efa/track.jpg<\/url><size>0<\/size><\/attachment><attachment id=\"dfa2848a-b93e-42ae-aac5-b3b39d65f4b6\" type=\"presentation/segment+preview\" ref=\"track:45210525-a5a8-4bab-a0a7-9451dfe8561e;time=T00:50:17:480F1000\"><mimetype>image/jpeg<\/mimetype><tags><tag>engage-download<\/tag><\/tags><url>https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/9724eb4c-9371-41a2-84b0-b216a65730fa/track.jpg<\/url><size>0<\/size><\/attachment><attachment id=\"f7bd3d95-f4c4-4779-9789-e7fc5f61cbdd\" type=\"presentation/segment+preview\" ref=\"track:45210525-a5a8-4bab-a0a7-9451dfe8561e;time=T01:12:50:680F1000\"><mimetype>image/jpeg<\/mimetype><tags><tag>engage-download<\/tag><\/tags><url>https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/cd276b03-f8c6-443b-b416-4d4ed7385e1b/track.jpg<\/url><size>0<\/size><\/attachment><attachment id=\"dcda8508-54f8-433b-9bfe-acddea561a27\" type=\"presentation/segment+preview\" ref=\"track:45210525-a5a8-4bab-a0a7-9451dfe8561e;time=T00:47:50:640F1000\"><mimetype>image/jpeg<\/mimetype><tags><tag>engage-download<\/tag><\/tags><url>https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/e966ff8e-ee9d-410a-bcb3-6f6f9d9de1c8/track.jpg<\/url><size>0<\/size><\/attachment><attachment id=\"2248f820-40ee-40b5-aad3-6431c963a921\" type=\"presentation/segment+preview\" ref=\"track:45210525-a5a8-4bab-a0a7-9451dfe8561e;time=T01:25:29:600F1000\"><mimetype>image/jpeg<\/mimetype><tags><tag>engage-download<\/tag><\/tags><url>https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/b812d83e-361b-4812-82cb-e092240809af/track.jpg<\/url><size>0<\/size><\/attachment><attachment id=\"570aa523-bc20-4b32-b423-903e3e149bc3\" type=\"presentation/segment+preview\" ref=\"track:45210525-a5a8-4bab-a0a7-9451dfe8561e;time=T00:12:37:160F1000\"><mimetype>image/jpeg<\/mimetype><tags><tag>engage-download<\/tag><\/tags><url>https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/3c970d74-059c-4d0a-af1d-feed5d3951f9/track.jpg<\/url><size>0<\/size><\/attachment><attachment id=\"8fad7d30-c681-47e4-bc0d-28e117fe58b0\" type=\"presentation/segment+preview\" ref=\"track:45210525-a5a8-4bab-a0a7-9451dfe8561e;time=T00:42:21:680F1000\"><mimetype>image/jpeg<\/mimetype><tags><tag>engage-download<\/tag><\/tags><url>https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/a1fe62b1-e6b6-4c49-9f7c-15131173bfac/track.jpg<\/url><size>0<\/size><\/attachment><attachment id=\"784fe4d6-c8c6-48ae-9860-adbedfafd945\" type=\"presentation/segment+preview\" ref=\"track:45210525-a5a8-4bab-a0a7-9451dfe8561e;time=T01:03:53:640F1000\"><mimetype>image/jpeg<\/mimetype><tags><tag>engage-download<\/tag><\/tags><url>https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/d5ecddad-75be-476a-9041-f408f91c7801/track.jpg<\/url><size>0<\/size><\/attachment><attachment id=\"0608259c-0d67-4022-bf57-6933b2fe870c\" type=\"presentation/segment+preview\" ref=\"track:45210525-a5a8-4bab-a0a7-9451dfe8561e;time=T00:00:42:800F1000\"><mimetype>image/jpeg<\/mimetype><tags><tag>engage-download<\/tag><\/tags><url>https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/76e6597e-d06c-462c-9370-55bbeb374af2/track.jpg<\/url><size>0<\/size><\/attachment><attachment id=\"623ba10e-6081-45d7-a8b8-74d4726f0221\" type=\"presentation/segment+preview\" ref=\"track:45210525-a5a8-4bab-a0a7-9451dfe8561e;time=T00:38:46:440F1000\"><mimetype>image/jpeg<\/mimetype><tags><tag>engage-download<\/tag><\/tags><url>https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/3166fdd2-7067-4d14-a973-178a413dd678/track.jpg<\/url><size>0<\/size><\/attachment><attachment id=\"4cc1f05f-97bc-4694-88f6-95706cf91e18\" type=\"presentation/segment+preview\" ref=\"track:45210525-a5a8-4bab-a0a7-9451dfe8561e;time=T00:40:50:440F1000\"><mimetype>image/jpeg<\/mimetype><tags><tag>engage-download<\/tag><\/tags><url>https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/72156a34-ffef-4961-9abf-8218c8d3f9f0/track.jpg<\/url><size>0<\/size><\/attachment><attachment id=\"882ae3b3-b461-4233-a9cb-09fd82856613\" type=\"presentation/segment+preview\" ref=\"track:45210525-a5a8-4bab-a0a7-9451dfe8561e;time=T01:27:46:720F1000\"><mimetype>image/jpeg<\/mimetype><tags><tag>engage-download<\/tag><\/tags><url>https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/bbb46923-b591-4330-ab8f-17a2e2d717c1/track.jpg<\/url><size>0<\/size><\/attachment><attachment id=\"a1786957-9780-4531-a73a-a4474ad1cc7b\" type=\"presentation/segment+preview\" ref=\"track:45210525-a5a8-4bab-a0a7-9451dfe8561e;time=T01:22:26:80F1000\"><mimetype>image/jpeg<\/mimetype><tags><tag>engage-download<\/tag><\/tags><url>https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/a1ae0961-a6b9-48ad-9f5f-eafce22f7656/track.jpg<\/url><size>0<\/size><\/attachment><attachment id=\"2a6a2641-e7fe-4c63-8b03-42458aa99da7\" type=\"presentation/search+preview\" ref=\"track:track-3\"><mimetype>image/jpeg<\/mimetype><tags><tag>engage-download<\/tag><\/tags><url>https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/attachment-3/track_1_000s_search.jpg<\/url><size>0<\/size><\/attachment><attachment id=\"3ee5c8ca-7909-42bd-b218-bd385285457f\" type=\"presentation/player+preview\" ref=\"track:track-3\"><mimetype>image/jpeg<\/mimetype><tags><tag>engage-download<\/tag><\/tags><url>https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/attachment-4/track_1_000s_player.jpg<\/url><size>0<\/size><\/attachment><attachment id=\"79f3e68a-2c1e-4520-ac9a-7141aaa1470a\" type=\"presentation/segment+preview\" ref=\"track:45210525-a5a8-4bab-a0a7-9451dfe8561e;time=T01:07:40:840F1000\"><mimetype>image/jpeg<\/mimetype><tags><tag>engage-download<\/tag><\/tags><url>https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/724f253a-afac-4c1b-b8c4-e01814af5a08/track.jpg<\/url><size>0<\/size><\/attachment><\/attachments><publications/><\/mediapackage>","mediaType":"AudioVisual","keywords":{"keywords":""},"modified":"2020-10-07T00:05:58+02:00","score":11.383356,"segments":{"segment":[{"index":0,"time":0,"duration":42800,"relevance":-1,"hit":false,"text":"IQ Science is to test crazy ideas Engineering is to put these ideas into Business 9HC 1 Andreas Holzinger 1 V0 709 043 Medical Informatics 30 W 30 11 2016 11 15 12 45 Lecture 06 Probabilistic Graphical 9 Models II From Bayesian Networks to 3 Graph Bandits a holzinger tugraz at Tutor markus plass student tugraz at http hci kdd org biomedical informatics big data new Group I mg 0 19 as","previews":{"preview":{"ref":"45210525-a5a8-4bab-a0a7-9451dfe8561e","$":"https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/d2128d70-db12-49ec-99e4-3f00f21cc51f/track.jpg"}}},{"index":1,"time":42800,"duration":71760,"relevance":-1,"hit":false,"text":"IU Agenda for today QHCI 06 Graph Bandits a very hot topic mz nglv Gvouv 1 709 549 as","previews":{"preview":{"ref":"45210525-a5a8-4bab-a0a7-9451dfe8561e","$":"https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/76e6597e-d06c-462c-9370-55bbeb374af2/track.jpg"}}},{"index":2,"time":114560,"duration":238600,"relevance":-1,"hit":false,"text":"TU Quiz QHCI cm l I I u N l l W n I I J E J O s 709 049 as","previews":{"preview":{"ref":"45210525-a5a8-4bab-a0a7-9451dfe8561e","$":"https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/274b400c-9e92-4b98-8b8a-52c3b852ad03/track.jpg"}}},{"index":3,"time":353160,"duration":144200,"relevance":-1,"hit":false,"text":"TU Quiz QHCI Gun l I I n M l 3 n I I J E 4 O u v i 1 u o u C 0 Nabr gImauv 5 m9 049 as","previews":{"preview":{"ref":"45210525-a5a8-4bab-a0a7-9451dfe8561e","$":"https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/d5e51fdc-4d7c-42ce-a3de-dae9ffdb8103/track.jpg"}}},{"index":4,"time":497360,"duration":58840,"relevance":-1,"hit":false,"text":"TU Quiz QHCI cm http sbcb bloch ox ac uk users ollver software Ho ur glv Gvouv e 709 5 9 as","previews":{"preview":{"ref":"45210525-a5a8-4bab-a0a7-9451dfe8561e","$":"https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/6e951bc3-5e85-4aa8-a0b2-5238edd24b13/track.jpg"}}},{"index":5,"time":556200,"duration":16720,"relevance":-1,"hit":false,"text":"Graph Mode M 01 Graphical Models and Decision laking","previews":{"preview":{"ref":"45210525-a5a8-4bab-a0a7-9451dfe8561e","$":"https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/3415228d-5667-4ed7-8379-69d95befa033/track.jpg"}}},{"index":6,"time":572920,"duration":10560,"relevance":-1,"hit":false,"text":"QHCI","previews":{"preview":{"ref":"45210525-a5a8-4bab-a0a7-9451dfe8561e","$":"https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/5891df7b-cac0-4cf9-b64f-d1049234811a/track.jpg"}}},{"index":7,"time":583480,"duration":150320,"relevance":-1,"hit":false,"text":"IU Decision Making Learn good policy for selecting actions QHCP Goal Learn an optimal policy for selecting best actions within a given context Fort 1 T 1 The world produces an uncertain context XI 6 X 2 The learner selects an action at e 1 t 3 The world reacts with a reward HUM E W Holzirglv Group 9 709 0 1905 Decision","previews":{"preview":{"ref":"45210525-a5a8-4bab-a0a7-9451dfe8561e","$":"https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/e8c5b1bf-d1de-4806-b6ac-a26482b5deac/track.jpg"}}},{"index":8,"time":733800,"duration":6480,"relevance":-1,"hit":false,"text":"QHCI","previews":{"preview":{"ref":"45210525-a5a8-4bab-a0a7-9451dfe8561e","$":"https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/15e202d1-27b0-496e-ac89-ca801fca8e2f/track.jpg"}}},{"index":9,"time":740280,"duration":16880,"relevance":-1,"hit":false,"text":"IU Key Challenges 9H0 Medicine is an extremely complex application domain dealing most of the time with uncertainties probable information When we have big data but little knowledge automatic ML can help to gain insight m 759 5 9 as","previews":{"preview":{"ref":"45210525-a5a8-4bab-a0a7-9451dfe8561e","$":"https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/15d297a9-96b7-41dd-894d-6e4fa92d1a84/track.jpg"}}},{"index":10,"time":757160,"duration":95200,"relevance":-1,"hit":false,"text":"LU Key Challenges 9H0 I Medicine is an extremely complex application domain dealing most of the time with uncertainties probable information I When we have big data but little knowledge automatic ML can help to gain insight I Structurae learning and prediction in large scale biomedical networks with probabilistic graphical models I If we have little data and deal with NP hard problems we still need the human in the loop m we 0 9 De","previews":{"preview":{"ref":"45210525-a5a8-4bab-a0a7-9451dfe8561e","$":"https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/3c970d74-059c-4d0a-af1d-feed5d3951f9/track.jpg"}}},{"index":11,"time":852360,"duration":148600,"relevance":-1,"hit":false,"text":"TU Example Why is prediction in proteins so important for us QHCI cm Noizirglv Gmuv u 709 049 De","previews":{"preview":{"ref":"45210525-a5a8-4bab-a0a7-9451dfe8561e","$":"https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/4be1e541-ff90-44b7-8f84-c116cccc87b1/track.jpg"}}},{"index":12,"time":1000960,"duration":33879,"relevance":-1,"hit":false,"text":"TU Recommended background reading cm v w Bishop C M 2007 Pattern Recognition and Machine Learnlng Heldelberg Sprunger Chapter 8 on graphical models openly available http research microsoft com en us um people cmblshop prml mmgu Group Murphy K P 2012 Machlne learnlng a probabilistic perspecllve MIT press Chapter 26 pp 907 Graphlcal model structure learnlng 11 QHCI Koller D Frledman N 2009 Probabmtlc graphical models princnples and techniques MIT press 709 549 as","previews":{"preview":{"ref":"45210525-a5a8-4bab-a0a7-9451dfe8561e","$":"https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/db18f079-68ae-447e-8e07-892c6773e8a2/track.jpg"}}},{"index":13,"time":1034839,"duration":124641,"relevance":-1,"hit":false,"text":"TU Three types of Probabilistic Graphical Models QHCI cm Undirected Markov random fields useful eg for computer vision Details Murphy 19 Directed Bayes Nets useful for designing models Details Murphy 10 y j 11 Factored useful for mlerence Iearmng I I I x II X Nolzirglr Group 13 m9 0 9 as","previews":{"preview":{"ref":"45210525-a5a8-4bab-a0a7-9451dfe8561e","$":"https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/540877d1-ab40-486f-a793-75be2b000005/track.jpg"}}},{"index":14,"time":1159480,"duration":6400,"relevance":-1,"hit":false,"text":"IU Factor Graphs learning at scaIe QHC What is the advantage of factor graphs Bayesian Networks Yes Somewhat Ancostra Gvnvrahw Process Markov Networks Yes No szm mummy mid F otm hal Factor Graphs No Yes U homt dismbuiod mforonco Tab 9 rec r ro Ra f Hemmch A hazow Noam Grove 14 m9 549 De","previews":{"preview":{"ref":"45210525-a5a8-4bab-a0a7-9451dfe8561e","$":"https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/a80a39c5-c445-4304-882f-6887423f6e96/track.jpg"}}},{"index":15,"time":1165880,"duration":48360,"relevance":-1,"hit":false,"text":"IU Factor Graphs learning at scale QHC What is the advantage of factor graphs Bayesian Networks Yes Somewhat Ancoetra Gvnvranvo Prorvss Markov Networks Yes No Lom Cont m s mm Potm hms Factor Graphs No Yes 7 an drst buif d mforonco Tab 9 rec t to RM Fero ch Amazo v mmgu Grove 4 m9 049 as","previews":{"preview":{"ref":"45210525-a5a8-4bab-a0a7-9451dfe8561e","$":"https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/ef3d6ebe-737c-42a6-aedd-9fc8ee781d11/track.jpg"}}},{"index":16,"time":1214240,"duration":41800,"relevance":-1,"hit":false,"text":"TU From structure to function prediction QHCI cm 2 0 Yowlosy 1 Slcondary semen Prodlctioll Primary an Contact and Snlvont Acctsilb lly Frodlctlon JD Struckuvo Com cl Map Baldi P Pollastrl G 2003 The prlncnpled design of large scale recursnve neural neiwork archntectures dag rnns and Ihe protein structure prediction problem The Journal of Machine Learning Research 4 575 602 Helzr gtrGrouv 15 709 my me","previews":{"preview":{"ref":"45210525-a5a8-4bab-a0a7-9451dfe8561e","$":"https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/f22280d2-d58d-4f7b-a45d-b93526d70fbe/track.jpg"}}},{"index":17,"time":1256040,"duration":62520,"relevance":-1,"hit":false,"text":"TU Protein Network Inference QHCI cm I Hypothesis most biological functions involve the interactions between many proteins and the complexity of living systems arises as a result of such interactions I In this context the problem of inferring a global protein network for a given organism I using all genomic data of the organism I is one ofthe main challenges in computational biology Yamanishi Y Vert J P Kanehisa M 2004 Protein network inference from multiple genomic data a supervised approach Biomformancs 20 Suppl 1 i363 I370 Holzir gli Group is m9 049 as","previews":{"preview":{"ref":"45210525-a5a8-4bab-a0a7-9451dfe8561e","$":"https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/a4fd407c-ab7b-4a5e-8434-7dcb1100fd5e/track.jpg"}}},{"index":18,"time":1318560,"duration":103600,"relevance":-1,"hit":false,"text":"TU Problem ls Graph lsomorphism NP complete 9H0 sun Borgwardt K M Ong C 5 Schonauer 5 1 31 s M Vishwanathan 5 Smola A J Kriegel H P A g 2005 Protein function prediction Via graph gs g 3 i O tum kernels Bioinformatics 21 Suppl 1 I47 I56 i imam mm 4min I Important for health informatics Discovering relationships between biological components I Unsolved problem in computer science I Can the graph isomorphism problem be solved in polynomial time I So far no polynomial time algorithm is known I It is also not known if it is NP complete O Holzirglv Group 7 759 5 9 as","previews":{"preview":{"ref":"45210525-a5a8-4bab-a0a7-9451dfe8561e","$":"https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/1336c585-1504-49a0-a7bf-0fc9572f8c67/track.jpg"}}},{"index":19,"time":1422160,"duration":55600,"relevance":-1,"hit":false,"text":"TU Example Protein Network Inference QHCI cm Protein network inference from multiple gag genomic data a supervised approach I I V 0 12 15 v J J NS i l lS i ch Il prrmunl Km Pl um mlcmcnnm Kg ILIK JIUJIIHIH _ KN I h lugcncllc prnlllc I Kup Kwl Km A n llnlcgmhmn nmznwcvoup 15 m9 549 De","previews":{"preview":{"ref":"45210525-a5a8-4bab-a0a7-9451dfe8561e","$":"https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/a2295e0a-f7ef-41fd-847b-40679ae28d5d/track.jpg"}}},{"index":20,"time":1477760,"duration":14760,"relevance":-1,"hit":false,"text":"TU Example Data fusion and Protein Annotation QHCI cm 3 _ r mm H m Lancknex G R De Ble T Crustlamnl N Jordan M Noble W S 2004 A statistical framework for genomic data fusnon Bloinformancs 20 16 2626 2635 Noxzmgu Gvouv 19 m9 0 1906","previews":{"preview":{"ref":"45210525-a5a8-4bab-a0a7-9451dfe8561e","$":"https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/6caf36e0-10ef-496a-adc7-2c4bc89ef2bd/track.jpg"}}},{"index":21,"time":1492520,"duration":28000,"relevance":-1,"hit":false,"text":"02 Bayesian Networks Bayes Nets","previews":{"preview":{"ref":"45210525-a5a8-4bab-a0a7-9451dfe8561e","$":"https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/f00a235f-714a-419a-b93b-f9c733c277ee/track.jpg"}}},{"index":22,"time":1520520,"duration":51480,"relevance":-1,"hit":false,"text":"TU Bayesian Network BN Definition QHCI cm I is a probabilistic model consisting of two parts I 1 a dependency structure and I 2 local probability models 1 p 1 1 n p I Pa 1 1 Where i ul arethe parents of BN nnerentlly niecel the uncertainty in the data We are a successla n arr age Demeen D csaa Iv meow and g aoh that al or to non El a 3 I dll Y e lSlD lal D osab ty C SI lDuIlO l n a sparse may on sea cn ng naepencenc elat ms in tne data Furthernrore tn 5 vrccel allows a fle ent strategies to integrate l C aata SOU CES Pearl J 1988 Probabilistic reasoning In Intelligent systems networks of plausrble Inference San Francisco Morgan Kaufmann u me 549 as um my Grove","previews":{"preview":{"ref":"45210525-a5a8-4bab-a0a7-9451dfe8561e","$":"https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/6d5a6667-c1ad-476b-9d18-56ec9de25081/track.jpg"}}},{"index":23,"time":1572000,"duration":32960,"relevance":-1,"hit":false,"text":"IU Example Directed Bayesian Network with 7 nodes 9H0 w j 2 V j 1 x v p _ _ n N V V V w V W V V V No mglv Gvouv 1 me 049 as","previews":{"preview":{"ref":"45210525-a5a8-4bab-a0a7-9451dfe8561e","$":"https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/e376250c-5db2-4426-b393-bbc4e7a2d19a/track.jpg"}}},{"index":24,"time":1604960,"duration":58360,"relevance":-1,"hit":false,"text":"TU Clinical Case Example cm gig Overmoyer B A Lee J M LerWIII M F 2011 Case 17 2011 A 49 Year Old Woman With a Mass In the Breast and Overlylng Skln Changes New England Journal of Medicine 364 23 2246 2254 Nolzwglv Gvouv","previews":{"preview":{"ref":"45210525-a5a8-4bab-a0a7-9451dfe8561e","$":"https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/519507a6-eec8-4782-b57d-9d09f351975e/track.jpg"}}},{"index":25,"time":1663320,"duration":57040,"relevance":-1,"hit":false,"text":"IU_ Important in Clinical practice prognosis QHCI the prediction ofthe future course of a disease conditional on the patient s history and a projected treatment strategy Danger probable Information Therefore valid prognostic models can be of great benefit for clinical decision making and of great value to the patient e g for notification and quality of life decisions Knaus W A Wagner D P Lynn Jr 1991 Short term mortality predictions for critically I hospitalized adults scnence and ethics Science 254 5030 389 Holzirglv Group 1 m9 0 19 as","previews":{"preview":{"ref":"45210525-a5a8-4bab-a0a7-9451dfe8561e","$":"https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/82285407-758b-480f-ae22-6a0e64912c2a/track.jpg"}}},{"index":26,"time":1720360,"duration":74560,"relevance":-1,"hit":false,"text":"TU Predicting the future on past data and present status QHCI cm t HI I CIII I ll It ll ltllt WI NINTH Itllt K RISK factors R sk factors Pathogenesns Pathogenesrs Disorders Disorders Pathophysrology Pathophysrology Flndlngs Findings Tests Treatments 0 pin ir mu past present future van Gerven MA 1 Taal B G Lucas P J F 2008 Dynamic Bayesnan networks as prognostic models for clinical patient management Journal ofBromedIcal Informatics 41 4 515 529 Narmgu Group 5 m9 5 9 as","previews":{"preview":{"ref":"45210525-a5a8-4bab-a0a7-9451dfe8561e","$":"https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/6446e2b3-f7a8-4e64-9df1-46d63281b45b/track.jpg"}}},{"index":27,"time":1794920,"duration":65560,"relevance":-1,"hit":false,"text":"TU Example Breast cancer Probability Table QHCI cm _l H w _r I r _r 1 w v mcr Wang X H et al 1999 Computer assured dragnOSIs of breast cancer usrng a data driven Bayesnan belief network International Journal of Medical Informatics 54 2 115 126 s 709 0 9 as","previews":{"preview":{"ref":"45210525-a5a8-4bab-a0a7-9451dfe8561e","$":"https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/c579ee9b-7e66-4ac9-931d-55243d9236bc/track.jpg"}}},{"index":28,"time":1860480,"duration":71800,"relevance":-1,"hit":false,"text":"TU Breast cancer big picture state of 1999 QHCI Gun Alcoholic 8 S n Smoking Thickening stchurgc Hormnncx Mcnupduu Pregnant Brcusl Cancer Ma Family Archllcclurul Ti uc Micmculu lhx lnr oDls tonion Asymmetry cutmnx Wang X H er al 1999 Computer assured dlagnOSIS of breast cancer usnng a data driven Bayesnan belief network International Journal of Medical Informatics 54 2 115 126 Nc rglvacuD 7 7m 0 1906","previews":{"preview":{"ref":"45210525-a5a8-4bab-a0a7-9451dfe8561e","$":"https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/83fb9e15-4f84-48c0-8ef9-0ac99e2300cc/track.jpg"}}},{"index":29,"time":1932280,"duration":58600,"relevance":-1,"hit":false,"text":"TU 10 years later Integration of microarray data QHCI am Integrating microarray data from multiple studies to Increase sample size approach to the development of more robust prognostic tests 3 N t I l Xu L Tan A Winslow R Geman D 2008 Merging mlcroarray data from separate breast cancer studies provndes a robust prognostic test BMC Blomformatlcs 9 1 125 139 No vaoup","previews":{"preview":{"ref":"45210525-a5a8-4bab-a0a7-9451dfe8561e","$":"https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/b29a2412-d7de-4f97-8434-0025dcb4a7a3/track.jpg"}}},{"index":30,"time":1990880,"duration":48840,"relevance":-1,"hit":false,"text":"TU Example Bayes Net with four binary variables QHCI cm Gene 1 Ptonl 0 8 Ptoffl 0 2 Gene 2 Gene 1 Gene 1 Gene 2 Gene 1 Gene 1 on 011 on 011 P1on1 O3 06 Ptonl 03 06 P101 07 04 P1011 O7 04 o Prognosns Gene 2 on Gene 2 on Gene 2 011 Gene 2 011 Gene 3 on Gene 3 of Gene 2 on Gene 3 011 Plgoodl 0 6 O 1 0 9 O 5 Plpoor O 4 O 9 O 1 O 5 Gevaen O Smet F D Tlmmerman D Moreau Y Moor B D 2006 Predicting Ihe prognosns of breast cancer by Integrating clinical and mncroarray daia wnth Bayesnan networksv Blomformancs 22 14 184 190 Ne1zr glmauD 9 m9 549 as","previews":{"preview":{"ref":"45210525-a5a8-4bab-a0a7-9451dfe8561e","$":"https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/5b258709-ba4f-46e0-aae6-49a61a64e25b/track.jpg"}}},{"index":31,"time":2039720,"duration":36320,"relevance":-1,"hit":false,"text":"TU Concept MarkowBlanket QHCI Gun Gevaert O Smet F D A Timmerman D Moreau Y Moor B D 2006 Predicting the prognosrs of breast O cancer by Integratmg J cllnlcal and mlcroarray data With Bayesran networks Blomformancs 22 14 O O 184 190 Nubr gerruup in m9 549 De","previews":{"preview":{"ref":"45210525-a5a8-4bab-a0a7-9451dfe8561e","$":"https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/39335258-3766-42e2-b74b-f21bfc4b5a12/track.jpg"}}},{"index":32,"time":2076040,"duration":24920,"relevance":-1,"hit":false,"text":"TU Dependency Structure first step 1 2 QHCI am I First the structure is learned using a search strategy I Since the number of possible structures increases super exponentially With the number of variables I the well known reed search al orithm K2 can be used In combination With the Bayesign Dirichlet BD scoring metric 1 r er r Nrk Nrk piSID xp r Ni Nij k r Ni k If I NM number of cases in the data set D havmg variable 1 in state associated With the i th instantiation of its parents in current structure 5 n is the total number of variables Holzr gerrouD 31 me 0490","previews":{"preview":{"ref":"45210525-a5a8-4bab-a0a7-9451dfe8561e","$":"https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/be112ef5-6009-4bc3-a476-4ecb3a9d77a0/track.jpg"}}},{"index":33,"time":2100960,"duration":7360,"relevance":-1,"hit":false,"text":"TU Dependency Structure first step 2 2 QHCI cm I Next rs calcu ated by summing over aH states of a varrabte I t _ t _ and have snmr ar meanrngs but refer to prror 0 knowledge for the parameters I When no khowtedge IS avarlab e they are estrmated usmg X r r I t I wrth Vthe equrva entsamp e srze I r thehumber of states ofvarrab ez and I L the number of mstantratrons of the parents of varrablez I It Icorrespondstothe gamma distribution I FmaHy pm t Is the prlor probablhty ot the structure I mMrs catcu ated by NW I II 11 I rll 1I 1 J I wrth 1 the number of parents of varlabte and u aH the varrables that are not a parent of x I Next pm In Is the probabrhty that there IS an edge from u to 7 whr e rm r rs the mverse e the probabrhty that there 15 no edge from u to Ir narmgu Grove a1 m9 5 9 as","previews":{"preview":{"ref":"45210525-a5a8-4bab-a0a7-9451dfe8561e","$":"https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/ea79344c-865a-4898-970a-5f1caead5f7a/track.jpg"}}},{"index":34,"time":2108320,"duration":43680,"relevance":-1,"hit":false,"text":"TU Parameter learning second step QHCI Gun Estlmatlhg the parameters of the local probablllty models correspondlng Wlth the dependency structure CPTs are used to model these local probablllty models For each varlable and mstantlatlon of lts parents there exlsts a CPT that conslsts of a set of parameters Each set of parameters was glven a uniform Dirichlet prlor 11 HH S DH LHHIN HI WM WWIZA VJH W l 7 cefmec as before 709 0 9 as Nol glv Grove 33","previews":{"preview":{"ref":"45210525-a5a8-4bab-a0a7-9451dfe8561e","$":"https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/f1acca10-7339-48f3-b8a8-45d235c6b0c9/track.jpg"}}},{"index":35,"time":2152000,"duration":52280,"relevance":-1,"hit":false,"text":"TU Predicting the prognosis of breast cancer integrated a QHCI cm mpmr H mv l Iv lunmrmmv l1 N Ulml V K Mm n I wnnml v wnuug T rp ngl u l VIIII JWI m u mm L4 um m1 mm um m mum w 2 4 u My m9 0 9 as u Hoh r glv Grove","previews":{"preview":{"ref":"45210525-a5a8-4bab-a0a7-9451dfe8561e","$":"https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/f9ba87ff-9ee0-4adc-a383-1b00646b7527/track.jpg"}}},{"index":36,"time":2204280,"duration":122160,"relevance":-1,"hit":false,"text":"LU Inference in Bayes Nets is intractable NP complete 9H0 I For certain cases it is tractable if I Just one variable is unobserved I We have singly connected graphs no undirected loops belief propagation I Assigning probability to fully observed set of variables I Possibility Monte Carlo Methods generate many samples according to the Bayes Net distribution and then count the results I Otherwise approximate solutions NOTE Sometimes it is better to have an approximate solution to a complex problem than a perfect solution to a simplified problem Nolzr gerrouD m9 5 19 as","previews":{"preview":{"ref":"45210525-a5a8-4bab-a0a7-9451dfe8561e","$":"https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/91059181-d401-4617-b548-81e0723871da/track.jpg"}}},{"index":37,"time":2326440,"duration":7480,"relevance":-1,"hit":false,"text":"3 Machine Learning on Graphs","previews":{"preview":{"ref":"45210525-a5a8-4bab-a0a7-9451dfe8561e","$":"https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/3166fdd2-7067-4d14-a973-178a413dd678/track.jpg"}}},{"index":38,"time":2333920,"duration":37360,"relevance":-1,"hit":false,"text":"I U_ Example Lymphoma is the most common blood cancer 9H0 The two main forms of lymphoma are Hodgkin lymphoma and non Hodgkin lymphoma NHL Lymphoma occurs when cells of the immune system called lymphocytes a type of white blood cell grow and multiply uncontrollably Cancerous lymphocytes can travel to many parts of the body including the lymph nodes spleen bone marrow blood or other organs and form a mass called a tumor The body has two main types of lymphocytes that can develop into lymphomas B lymphocytes B cells and T lymphocytes T cells www lymphomaorg http imagebankhematologyorg Holzir gtv Swan 37 709 0 19 06","previews":{"preview":{"ref":"45210525-a5a8-4bab-a0a7-9451dfe8561e","$":"https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/9af7c54f-3da7-49c3-852e-bf3b66f6aded/track.jpg"}}},{"index":39,"time":2371280,"duration":79160,"relevance":-1,"hit":false,"text":"TU ML tasks on graphs QHCI Gun Discover unexplored interactions in PPI networks and gene regulatory networks Learn the structure Reconstruct the structure _ 7 1 Dlnrvch M T Klau G W Rosenwald A Dandekar T Muller T 2008 Idenhfymg functional modules In protein protein lnteractxon networks an Integrated exact approach Blomformatlcs 24 13 l223 l231 Holzlrglr Gvouu BS 709 5 9 as","previews":{"preview":{"ref":"45210525-a5a8-4bab-a0a7-9451dfe8561e","$":"https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/9ccda672-3ca8-40f7-a797-86c4ccc1daf6/track.jpg"}}},{"index":40,"time":2450440,"duration":23880,"relevance":-1,"hit":false,"text":"TU From structure to function QHCI cm _ _ love I J A Protocol for Computer BJsed Protein Structure and runcnon Predmmn mun 0 _u EEK I lvv CKAIVY C V Umzi http www10ve com video 3259 a protocoHor computer based proteln structure tunctvon new 5mm 39 me 049 as","previews":{"preview":{"ref":"45210525-a5a8-4bab-a0a7-9451dfe8561e","$":"https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/72156a34-ffef-4961-9abf-8218c8d3f9f0/track.jpg"}}},{"index":41,"time":2474320,"duration":67360,"relevance":-1,"hit":false,"text":"TU Interesting Hubs tend to link to small degree nodes QHCI cm Nodes proteins Links physical interactions binding Puzzling pattern Hubs tend to link to small degree nodes Why IS this puzzling in a randcm r et xcrk the meta it ma 3 noce x 1 1 degree l 5 to a nose nit egree lS AA I p 015 p 0001 Jeong H Mason S P Barabasn A L Oltvai Z N 2001 Lethality and centrality In protein networks Nature 411 6833 41 42 709 049 as AU mimgu Grouv","previews":{"preview":{"ref":"45210525-a5a8-4bab-a0a7-9451dfe8561e","$":"https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/a0c59d71-959f-4c49-ab51-5e462bf4c221/track.jpg"}}},{"index":42,"time":2541680,"duration":9680,"relevance":-1,"hit":false,"text":"TU Interesting Hubs tend to link to small degree nodes QHCI cm Nodes proteins Links physical interactions binding Puzzling pattern Hubs tend to link to small degree nodes Why IS this puzzling in a anacm net usrh tr e Drcbab it ma a nod eg ee l x510 a node x 1 egree ls i M I I 2 3 453 25 p l 1 p 0 mm Jeong H Mason S P Barabasn A L Oltvai Z N 2001 Lethality and centrality In protein networks Nature 411 6833 41 42 mg 0 1un a elm 5mm","previews":{"preview":{"ref":"45210525-a5a8-4bab-a0a7-9451dfe8561e","$":"https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/a1fe62b1-e6b6-4c49-9f7c-15131173bfac/track.jpg"}}},{"index":43,"time":2551360,"duration":39560,"relevance":-1,"hit":false,"text":"TU Example Subgraph Discovery QHCI cm MIMIJI V K 4 l4 I M Um z ml WWW r w w it w s_ n ppm huki twqr a 1 NR HGQ JM MW mum wmvcyxw W 1111 NM HRH H ul l Nehr glvaouD 41 m9 5 905","previews":{"preview":{"ref":"45210525-a5a8-4bab-a0a7-9451dfe8561e","$":"https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/f0952a7a-d650-4656-90f4-bb7e7dc5e949/track.jpg"}}},{"index":44,"time":2590920,"duration":5280,"relevance":-1,"hit":false,"text":"TU Example Subgraph Discovery QHCI cm do saw m h String My HIGH ENERGY PHYS CS THEORY ASTROPHYSICS Hm M W W Model 01 um Anbotfopy MMNMM An Alternative To Compacmication lesnmaled bndgeness 1276 GENERAL RELAYIVITY AND QUANTUM COSMOLOGY K x uh n M mu um mm lwau mvlJppwg A mum mu mmm AWMHW l vlsmvkx Lue vml gx a n maSnnIExv Dimnuon HIGH ENERGY PHYS CS Ndmml m hh lll 4 m mu v x PHENOMENOLOGV 11th l in 14 H m mums Gvoup u 709 0 1905","previews":{"preview":{"ref":"45210525-a5a8-4bab-a0a7-9451dfe8561e","$":"https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/e9325995-b460-4c65-9c8a-72f0e3972b25/track.jpg"}}},{"index":45,"time":2596200,"duration":152200,"relevance":-1,"hit":false,"text":"TU Why do we want to apply ML to graphs QHCI cm I A Discovery of unexplored interactions I 8 Learning and Predicting the structure I C Reconstructing the structure I Which joint probability distributions does a graphical model represent I How can we learn the parameters and structure of a graphical model u m w m The chemical space O possmle small or ganic molecules 3 w stars In the observ able universe Nolzmglv 6mm 41 m9 on as","previews":{"preview":{"ref":"45210525-a5a8-4bab-a0a7-9451dfe8561e","$":"https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/e62c918e-485c-45c6-9e7f-742c7e1446c4/track.jpg"}}},{"index":46,"time":2748400,"duration":39920,"relevance":-1,"hit":false,"text":"TU Example Question Predicting Function from Structure QHCI cm 11 my I I vasmwuuvxm magnum mm um I vasmmvxm magnum mam um I SA vaounuLI1I Hal 4 VIVIXGKLDILI It xm v I am x I mun cu r I m1 x n not n x v V I nu max I Inuuuru mm I m1 I u 19me I n 2a 2 O n z 4 How similar are two graphs How similar is their structure How similar are their node and edge labels LHKLl M KAI hill A l 1mm ll Kn ll lllp i Lil Illtn i v l l H Hi llllh Jr hmi l 1 Vi i vl lr m mtunui mm i Ilhi rn mitrawmmwy nu ms nu I m Noizr glmauD 13 m9 549","previews":{"preview":{"ref":"45210525-a5a8-4bab-a0a7-9451dfe8561e","$":"https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/6ec9c41f-f1dd-41d8-966c-cc852d005075/track.jpg"}}},{"index":47,"time":2788320,"duration":82320,"relevance":-1,"hit":false,"text":"TU Graph Comparison QHCI cm I Similar Property Principle Molecules having similar structures should have similar activities I Structure based representations Compare molecules by comparing substructures e g I Sets as vectors Measure similarity by the cosine distance I Sets as sets Measure similarity by the Jaccard distance I Sets as points Measure similarity by Euclidean distance I Problems Dimensionality Non Euclidean cases Noimgu Group 4 m9 049 as","previews":{"preview":{"ref":"45210525-a5a8-4bab-a0a7-9451dfe8561e","$":"https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/ffac8015-7952-4061-a9c5-00c95b5de942/track.jpg"}}},{"index":48,"time":2870640,"duration":16880,"relevance":-1,"hit":false,"text":"4 Little Excursus What is similarity","previews":{"preview":{"ref":"45210525-a5a8-4bab-a0a7-9451dfe8561e","$":"https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/e966ff8e-ee9d-410a-bcb3-6f6f9d9de1c8/track.jpg"}}},{"index":49,"time":2887520,"duration":56160,"relevance":-1,"hit":false,"text":"TU What is Similar QHCI cm Image credlt tooEamonn Keogh 2008 Ne pr glv Gvouo 16 709 on as","previews":{"preview":{"ref":"45210525-a5a8-4bab-a0a7-9451dfe8561e","$":"https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/dffc5f4a-3b7d-4af9-9753-04151aaca684/track.jpg"}}},{"index":50,"time":2943680,"duration":55440,"relevance":-1,"hit":false,"text":"TU QHCI Gru i M m I M M M mu 4 N wm rn 1 41er nxlwm mm mwm t mmw Noungucveup 47 m9 0 as","previews":{"preview":{"ref":"45210525-a5a8-4bab-a0a7-9451dfe8561e","$":"https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/6dfc7169-200b-4084-b780-bfd15c6e3651/track.jpg"}}},{"index":51,"time":2999120,"duration":18360,"relevance":-1,"hit":false,"text":"TU QHCI cm M vam A M 4 IHIH M 11 l M 5 w 0 Mum m u h wrm U mm mm mum Mm x t mmw vaouD 4s me 5 9 as","previews":{"preview":{"ref":"45210525-a5a8-4bab-a0a7-9451dfe8561e","$":"https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/c342a425-cabb-40a0-a732-535300542d22/track.jpg"}}},{"index":52,"time":3017480,"duration":88000,"relevance":-1,"hit":false,"text":"TU Similarity and Correspondence QHCI cm lnn V l A J W H I l J M li i i JW X NL HWILJI rmvmvr Mum mmwumx Mun me mr e http www lnf usi ch bronstein Structure Structure Neizr glvaouD 49 m9 549 as","previews":{"preview":{"ref":"45210525-a5a8-4bab-a0a7-9451dfe8561e","$":"https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/9724eb4c-9371-41a2-84b0-b216a65730fa/track.jpg"}}},{"index":53,"time":3105480,"duration":19400,"relevance":-1,"hit":false,"text":"TU Invariant Similarity QHCI cm Transformanon naimgu Gvouv so 709 049 as","previews":{"preview":{"ref":"45210525-a5a8-4bab-a0a7-9451dfe8561e","$":"https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/53f08c19-ebc5-4ce7-9ddb-2681eb42c00c/track.jpg"}}},{"index":54,"time":3124880,"duration":87320,"relevance":-1,"hit":false,"text":"I Gromov Hausdorff dist finding the opt correspondence Gromov M 1984 infinite groups as geometric objects Felix Hausdorff Michall Gromov 1868 1942 1943 w Ym y Metric space Metric space 1 IGM V Y min max r r n 1 u g i _ i t r lh S tr r LY I h Li S t I 2 L Discrete optimization over correspondences is NP hard Neimgu Group 51 m9 0 1906","previews":{"preview":{"ref":"45210525-a5a8-4bab-a0a7-9451dfe8561e","$":"https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/d6759621-7816-423d-9969-dd4cccb1b4ba/track.jpg"}}},{"index":55,"time":3212200,"duration":70480,"relevance":-1,"hit":false,"text":"TU Example QHCI cm 0 Sormanv C 2010 How Riemanman Manlfokis Converge A Survey arXIv preprml awn 1006704117 mums Gvouv 5 me my 05","previews":{"preview":{"ref":"45210525-a5a8-4bab-a0a7-9451dfe8561e","$":"https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/8c57954f-4451-45f0-98ba-131bb2f55272/track.jpg"}}},{"index":56,"time":3282680,"duration":9280,"relevance":-1,"hit":false,"text":"5 Probabilistic Topic Models","previews":{"preview":{"ref":"45210525-a5a8-4bab-a0a7-9451dfe8561e","$":"https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/5f67c6bd-1941-4052-a696-96aa17daefbc/track.jpg"}}},{"index":57,"time":3291960,"duration":102080,"relevance":-1,"hit":false,"text":"TU Topic modelling small topic but hot topic in ML QHCI cm TOPIC MODELING PROBABILISTIC MODELING STATISTICS MACHINE LEARNING DATA SCIENCE mg 549 De Nolzwglv Gvouv 54","previews":{"preview":{"ref":"45210525-a5a8-4bab-a0a7-9451dfe8561e","$":"https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/35a92b3f-2e1e-4308-9c90-0a2b39f6bb51/track.jpg"}}},{"index":58,"time":3394040,"duration":57880,"relevance":-1,"hit":false,"text":"TU Geometry of Topic Models QHCI cm I l urd I Q loplc O Uh cn cd documcnl 8 gcncrulcd ducumcnl I mrdlh I ord Blei D M 2012 Probabilistic topic models Communications of the ACM 55 4 77 84 d0I 10 1145 2133806 2133826 Noungucveup 55 m9 049 De","previews":{"preview":{"ref":"45210525-a5a8-4bab-a0a7-9451dfe8561e","$":"https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/eee918ba-be51-4bd6-97a0-ab7f740bd4c7/track.jpg"}}},{"index":59,"time":3451920,"duration":131960,"relevance":-1,"hit":false,"text":"TU Generative statistical model for natural language QHCI cm Seelung Lile s Bare Genetic Necessities H__ J Given the parameters a and E the joint distribution of a topic mixture 9 a set of N topics Z and a set of N words w is given by 9 L 1 15 M Win i311i9 Wu 3 III 0 Blei D M Ng A Y Jordan M 2003 Latent dirichlet allocation The Journal of machine Learning research JMLR 3 993 1022 Hoizr guGrouD 55 m9 0 1906","previews":{"preview":{"ref":"45210525-a5a8-4bab-a0a7-9451dfe8561e","$":"https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/57a64a0d-b0dc-4640-8237-a4df5ef3f941/track.jpg"}}},{"index":60,"time":3583880,"duration":102680,"relevance":-1,"hit":false,"text":"TU Motivation to get insight into unknown document sets QHCI cm http ago dst glthub Io dfr browser demo N modeI scaled Noizr glvaouD 57 m9 M906","previews":{"preview":{"ref":"45210525-a5a8-4bab-a0a7-9451dfe8561e","$":"https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/de1e857b-3297-4b8e-8197-4734ae81cf41/track.jpg"}}},{"index":61,"time":3686560,"duration":71400,"relevance":-1,"hit":false,"text":"TU Example from Bioinformatics QHCI cm A Funcmrul moon 1 Fur mom mod m 2 arm Emm Emmv 2 Er zr w 3 Emm 4 1 Ga J I l Em am m 3 I Yogic 1 Yo 2 P nlmr 1 WW Llnmbom pairg O B mama mmm 9mm C sum Imam n Konieuny S G Diexz L Mchardy A C 2011 Inferring functional modules of proteln famllles wnth probabilistic oplc models BMC blonnformancs 12 1 1 Nmzngusmup 55 mg 0 19","previews":{"preview":{"ref":"45210525-a5a8-4bab-a0a7-9451dfe8561e","$":"https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/661b3521-6399-4131-b8d2-bfd5c1d47441/track.jpg"}}},{"index":62,"time":3757960,"duration":27640,"relevance":-1,"hit":false,"text":"TU Eval scheme for inferred potential functional modules QHCF cm 06 am an 064 an ac ac on m 4 0 1 on or Hg 0 on ocv x wan y H I J x MJuImy A x 1m Ir m v g mn mun mum w vlrw amulm mv wuuumn m mum qu muu mum 2 m 1 Nc rglvaouD 59 m9 5 905","previews":{"preview":{"ref":"45210525-a5a8-4bab-a0a7-9451dfe8561e","$":"https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/17e75605-f22c-49d6-9fbc-c0f4900bcfbc/track.jpg"}}},{"index":63,"time":3785600,"duration":48040,"relevance":-1,"hit":false,"text":"TU Example QHCI am Toplc proportions and T m n epics Docu e 5 ass gnmenls Seeking Life s Bare Genetic Necessities 2 We only observe the docs the other structure is hidden then we compute the posterior p t p a docs Hcizrg vaouv so my 049 De","previews":{"preview":{"ref":"45210525-a5a8-4bab-a0a7-9451dfe8561e","$":"https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/82b5188e-6bfa-463f-a815-e6bfc091c6eb/track.jpg"}}},{"index":64,"time":3833640,"duration":34960,"relevance":-1,"hit":false,"text":"IU_ We find topics through a Generative Probabilistic Model 9H0 Goal to get insight in unknown document collections See a nice demo http agoldst github Io dfr browser demo Wmodel gnd Topic pmpomons and TODICS Documenls asstgnmenls Seeking Live s Bare Genetic Necessities V z X gt 3 5 i Each doc is a random mix of corpus wide topics and each word is drawn from one of these topics Noimgu Giana si 759 on De","previews":{"preview":{"ref":"45210525-a5a8-4bab-a0a7-9451dfe8561e","$":"https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/d5ecddad-75be-476a-9041-f408f91c7801/track.jpg"}}},{"index":65,"time":3868600,"duration":8800,"relevance":-1,"hit":false,"text":"TU Output Example 4 learned topics details in Blei 2008 QHCI cm llumnn mnlnlinn llM ilM Hlllllllll l gl llnnn vmlutimuu y In Illml l Inn Ill l Inll ll l lzl inl nl nmliull gnu lil ul gnlli lls mum s lulu gt lll s lll t l l si zlllt v ulll llll l M qlu nu origin Inll ll l lill _ ll lll 34mm llll nluuy 111 1le u k Illnlwlllnr grmu lmiI1 y lvm Mum min pln lngvnvlix rumml lIllIlll l map living inlwtirgb pnmlll l inlurnmliun xli 1 il malaria llll l lml gt m Iim gmup I ll l ll Ilt l l k mapping nvw ll il lll wl lwnn prnjm t l u llllllHl 11 wqut lnw Hllllllnll llllll l l lll l l simululiuns Hal glv Evans 6 m9 L349 05","previews":{"preview":{"ref":"45210525-a5a8-4bab-a0a7-9451dfe8561e","$":"https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/f0de732e-5ddd-4caf-ab64-ba2e7939cb70/track.jpg"}}},{"index":66,"time":3877400,"duration":17880,"relevance":-1,"hit":false,"text":"TU Output Example 4 learned topics details in Blei 2008 QHCI cm llumun mnlnlinn lllN ilN unnpnlvl gummv ulutinnulj llmt Inmlvl Inn Ill l lml ll l lzl inllu umliull ngIl lil ul gilllisllls mum s lulu gt llm lll t l l sl zllll v llllllllll l h N qllt nu ul iglll Inll ll l lill _ l lll gvnv llll Ill my 111 1le 1 k Illnlm lllnr grmq ll illlls x slvnh l quvnri11g pln lngvnvlit t nllll nl llllllll l map living illl wtium pnrnllvl inlnrmnliun xli r il Illulul izl lnvlllmls ut m I i gm mp l ll 1 il m tm rk Innppillu nvw lmrmitlw wl nmn j l l u llllilwl 11 N qllt llt l s l HllllllHll llllN l l lllH l simululimls mum Evans 51 m9 L349 us","previews":{"preview":{"ref":"45210525-a5a8-4bab-a0a7-9451dfe8561e","$":"https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/addb0b35-fee3-4ef9-9680-e9d913c71982/track.jpg"}}},{"index":67,"time":3895280,"duration":13600,"relevance":-1,"hit":false,"text":"TU LDA is an example for a probabilistic graphical model QHCI cm V v V V v s a I no L n H r p i I K Encodes assumptions on data With a factorization of the Jomt Connects assumptions to algorithms for computing With data Defines the posterior through the pint HolzngliGiouD 53 m9 04905","previews":{"preview":{"ref":"45210525-a5a8-4bab-a0a7-9451dfe8561e","$":"https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/ca1ad350-3192-41f2-b2be-734b1bba7373/track.jpg"}}},{"index":68,"time":3908880,"duration":40000,"relevance":-1,"hit":false,"text":"TU LDA is an example for a probabilistic graphical model QHCI cm u it in n H u p i I K Encodes assumptions on data With a factorization of the iomt Connects assumptions to algorithms for computing With data Defines the posterior through the pint Nolzr gliGiouD 53 709 54905","previews":{"preview":{"ref":"45210525-a5a8-4bab-a0a7-9451dfe8561e","$":"https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/835ca2af-7235-4911-aa10-c3442fcbf6a6/track.jpg"}}},{"index":69,"time":3948880,"duration":20480,"relevance":-1,"hit":false,"text":"IU Posterior inference QHCI I H l n Ill g 7 ri 0 z w l 0 it l 0 z w We can t compute the denominator the marginal p w therefore we use approximate inference h 0 zl However this do not scale well Noimgu new M m9 549 as","previews":{"preview":{"ref":"45210525-a5a8-4bab-a0a7-9451dfe8561e","$":"https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/9c3c7f8a-c597-4d07-bc3b-62bae8b79223/track.jpg"}}},{"index":70,"time":3969360,"duration":42480,"relevance":-1,"hit":false,"text":"TU For big data stochastic variational inference QHCI cm Hoffman M D Blei 0 M Wang C Paislev J 2013 Stochastic variational inference The Journal of Machine Learning Research 14 1 1303 1347 NeizrglvaouD as 709 on us","previews":{"preview":{"ref":"45210525-a5a8-4bab-a0a7-9451dfe8561e","$":"https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/33c4b33a-8a0a-45f8-9086-354b82a36b26/track.jpg"}}},{"index":71,"time":4011840,"duration":23480,"relevance":-1,"hit":false,"text":"IU Stochastic variational inference in LDA QHCI 44 Wu H v v v Li H U Ji U z r DIR UV I A o i Sammie a oocumem Estimaie me ocai war a O ai parawete s JS 9 We CLANOHI topics FO V HWIO WCO GTO ICO CS VO W TT OSC OCH oa av eters poem Upcate topics as a eg 10d average 0 wormed me are current Ionics NuiznglvaouD as m9 549 De","previews":{"preview":{"ref":"45210525-a5a8-4bab-a0a7-9451dfe8561e","$":"https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/68c1787c-db47-4066-80ee-185a2205586c/track.jpg"}}},{"index":72,"time":4035320,"duration":8360,"relevance":-1,"hit":false,"text":"TU For big data stochastic variational inference QHCI cm Hoffman M D Blei Dr M Wang C Paislev J 2013 Stochastic variational inference The Journal of Machine Learning Research 14 1 1303 13477 Ncizr gquvoup es 709 0 9 as","previews":{"preview":{"ref":"45210525-a5a8-4bab-a0a7-9451dfe8561e","$":"https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/9abd0d03-0891-458b-8006-0a911695502d/track.jpg"}}},{"index":73,"time":4043680,"duration":17160,"relevance":-1,"hit":false,"text":"IU Stochastic variational inference in LDA QHCI u if H v v v Li H izm a r 3 U I A O i Samo e a Document ESIHWBIC 10 OCQ JEN 21 mai C8 Z J C C S iSi 10 CLM CHI 00 CS Pow interrveo ate Ionics rO n r ose ioca Caraweiers me Upcate topics as a we gr ted average 0 mowed are are curren 10 cs Noizr glvaouD as 7591 4905","previews":{"preview":{"ref":"45210525-a5a8-4bab-a0a7-9451dfe8561e","$":"https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/1e052c0a-5870-4dbe-ad1b-3c15715ccbd2/track.jpg"}}},{"index":74,"time":4060840,"duration":27080,"relevance":-1,"hit":false,"text":"TU Stochastic variational inference QHCI cm lmiuiil andumh cl m iCP xll hcdulc lppluprmlk h rvpul nnpic i dmumcni L umiumih Imxn Ihc LILI Ci lmluii c g 1 In v 1 A rvpul M X I J _ unlil liui mmxncmx i JUL 4 wnxnuc m 1 A 1 cl mmnmimc Alj hx i 1r l 17 i A unlil lurcwi Hoffman M D Blei D M Wang C Paislev J 2013 Stochastic variational inference The Journal of Machine Learning Research 14 1 1303 1347 vamm 57 ms 049 as","previews":{"preview":{"ref":"45210525-a5a8-4bab-a0a7-9451dfe8561e","$":"https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/724f253a-afac-4c1b-b8c4-e01814af5a08/track.jpg"}}},{"index":75,"time":4087920,"duration":35200,"relevance":-1,"hit":false,"text":"IU Approximate inference can be difficult to achieve QHCI KNOWLEDGE v v Make assumptions Discover patterns Predict 81 Explore Q 1 I tn Approximate inference can be difficult to derive Especially true for models that are not conditionally conjugate Discrete choice modelst Bayesian generalized linear models Holds us back from trying many models Holzr gerrouD as me 5 1906","previews":{"preview":{"ref":"45210525-a5a8-4bab-a0a7-9451dfe8561e","$":"https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/f16d5787-f7b0-4635-8675-8ea5c56411c1/track.jpg"}}},{"index":76,"time":4123120,"duration":37679,"relevance":-1,"hit":false,"text":"TU Black Box Approach QHCI cm REUSABLE MASS VE VARIATIONAL DATA FAMILIES ANY MODEL BLACK BOX _ VARIATIONAL INFERENCE O Easily use variational inference with any model No exponential family requirements No mathematical work beyond specifying the model Nolzr glmauD 59 m9 5 9 as","previews":{"preview":{"ref":"45210525-a5a8-4bab-a0a7-9451dfe8561e","$":"https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/13803432-0117-4425-9cca-21ef610c3271/track.jpg"}}},{"index":77,"time":4160799,"duration":164201,"relevance":-1,"hit":false,"text":"IU_ Conclusion What are future challenges 9 Flexible and expressive components for building models are of utmost importance Scalable and generic inference algorithms multi task and transfer learning Usability gets a totally new importance Easy to use algorithms for the non expert user to stretch probabilistic modeling into new areas Topic models are one approach towards detection of topics in document collections I Example Identifying re occurring patterns in such data collections gaining new knowledge Nolzr gerrouD 70 m9 mg as","previews":{"preview":{"ref":"45210525-a5a8-4bab-a0a7-9451dfe8561e","$":"https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/4cef5745-e19a-4571-9b11-84b3948d8d8e/track.jpg"}}},{"index":78,"time":4325000,"duration":45680,"relevance":-1,"hit":false,"text":"6 Graph Bandits","previews":{"preview":{"ref":"45210525-a5a8-4bab-a0a7-9451dfe8561e","$":"https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/d7ec3d3e-0242-4bb6-ae8e-10e8cc9d8b47/track.jpg"}}},{"index":79,"time":4370680,"duration":29720,"relevance":-1,"hit":false,"text":"TU The complexities of optimization S basitien Bubeck QHCI am l m a bandit Bubeck S Cesa Bianchi N 2012 Regret Analysns of Stochastic and ORF523 The complexmes of Nonstochastic Multi _ _ armed Bandit optimization Problems Machine Learning 5 1 1 122 hnps blogs pnncetonedu imabandit Also very Interesting Bubeck S 2015 Convex optuzanon Algorithms and complexity Foundations and Trends in Machine Learning 8 134 231 357 Newgu 6mm 7 759 on us","previews":{"preview":{"ref":"45210525-a5a8-4bab-a0a7-9451dfe8561e","$":"https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/cd276b03-f8c6-443b-b416-4d4ed7385e1b/track.jpg"}}},{"index":80,"time":4400400,"duration":40200,"relevance":-1,"hit":false,"text":"6 Graph Bandits","previews":{"preview":{"ref":"45210525-a5a8-4bab-a0a7-9451dfe8561e","$":"https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/63a39edf-fc5e-43b7-8b22-dad80bdda431/track.jpg"}}},{"index":81,"time":4440600,"duration":253640,"relevance":-1,"hit":false,"text":"IU What is a bandit 9 I Slot machine bandit robs your money I One armed bandit I Very simple model for sequential decision making under uncertainty I Main challenge exploration versus exploitation I Many application domains A B Testing Crowdsourcing optimization search Nclzmgu Grouv 73 m9 049 as","previews":{"preview":{"ref":"45210525-a5a8-4bab-a0a7-9451dfe8561e","$":"https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/e2acd983-4a95-424d-8ad9-73111a1d8178/track.jpg"}}},{"index":82,"time":4694240,"duration":126800,"relevance":-1,"hit":false,"text":"TU Multi Armed Bandits problem QHCI om I Multi armed bandit a gambler strategically operating multiple machines in order to draw the highest possible profits I There are n slot machines einarmige Banditen I Each machine 1 returns a reward y z P y 9 I Challenge The machine parameter 0 is unknown I Which arm of a slot machine should a gambler pull to maximize his cumulative reward over a sequence of trials stochastic setting or adversarial setting Nolzr glmauv 7 759 549 as","previews":{"preview":{"ref":"45210525-a5a8-4bab-a0a7-9451dfe8561e","$":"https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/3016dfa2-6f63-4838-905a-56503c1ddea3/track.jpg"}}},{"index":83,"time":4821040,"duration":58200,"relevance":-1,"hit":false,"text":"IU Machine Parameters of the k armed Bandit QHCI Each arm a either wins reward 1 with fixed unknown probability 1 or loses reward 0 with fixed unknown probability 1 1 All draws are independent given ill m Problem How to pull arms to maximize the total reward O Nclzir gli Evans 75 m9 049 as","previews":{"preview":{"ref":"45210525-a5a8-4bab-a0a7-9451dfe8561e","$":"https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/98f14e9e-0bd5-4814-accc-27dd0c253161/track.jpg"}}},{"index":84,"time":4879240,"duration":66840,"relevance":-1,"hit":false,"text":"TU Underlying Principle of the k Armed Bandits problem 9H0 sun I Let 1 E l n be the cltoice of a machine at time r I Let y E E be the outcome With a mean of hi I Now the given policy maps all history to a new chOIce I i i uh p ii I The problem Find a policy n that m iX y I Now two effects appear when choosing such machine I You collect more data about the machine knowledge I You collect reward I Exploration and Exploitation I Exploration Choose the next action Ll to miiiilli Ii H I Exploitation Choose the next action it to iiiuxiv I models an agent that simultaneously attempts to acquire new knowledge called exploration and optimize his or her deCISIons based on eXIsting knowledge called eprOitation The agent attempts to balance these competing tasks in order to maximize total value over the period of time considered More information http researchmicrosoft com en us proiects bandits Nolzr glvaouD 75 m9 0 9 as","previews":{"preview":{"ref":"45210525-a5a8-4bab-a0a7-9451dfe8561e","$":"https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/2d278a9a-1a23-449f-8cdd-718ef36b6bf5/track.jpg"}}},{"index":85,"time":4946080,"duration":99400,"relevance":-1,"hit":false,"text":"TU MAP Principle Optimism in the face of uncertainty QHCI cm 1 Itm xtlwv M 11 1HH M _ EM the higher the estimated the higher the W reward the higher the chance uncertainty the higher the to select the action chance 0 select the action Auer P Cesa Blanchl N Fischer P 2002 Flnlte time analysns of the multlarmed bandit problem Machlne learning 47 2 3 235 256 Nob mglvaouD 77 759 049 as","previews":{"preview":{"ref":"45210525-a5a8-4bab-a0a7-9451dfe8561e","$":"https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/a1ae0961-a6b9-48ad-9f5f-eafce22f7656/track.jpg"}}},{"index":86,"time":5045480,"duration":23400,"relevance":-1,"hit":false,"text":"TU A bandit in a graph is still a bandit QHCI cm 0 Let G a mown graph With K nodes 11 2 K Lc t f be 3 unknown function defined on the set 0 nodes oFort lton 5 w 1 may Obsuu usuard i 1 o 603 mmimizc sum of apt ted rewards Equwaientiv minimize regret R ir r u 1 where 7 mm A Mi 0 C Care about the Case when H n Ncizmglv Gvouo 73 ms 549 De","previews":{"preview":{"ref":"45210525-a5a8-4bab-a0a7-9451dfe8561e","$":"https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/1b3fc3a1-9d79-42f4-9d9c-9809b2121efa/track.jpg"}}},{"index":87,"time":5068880,"duration":24200,"relevance":-1,"hit":false,"text":"TU Smooth Graph Function QHCI cm Nobr glvaouD 79 759 my me","previews":{"preview":{"ref":"45210525-a5a8-4bab-a0a7-9451dfe8561e","$":"https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/628b2a11-514b-4fa3-8ba9-9aca78ad4e2e/track.jpg"}}},{"index":88,"time":5093080,"duration":6560,"relevance":-1,"hit":false,"text":"TU Knowledge Representation in MAB QHCI cm Knowledge can be represented in two ways l as fullhistory f l l 1 7 V 1 or 2 as belief 1 Hi H I where 0 are the unknown parameters c6 all machines The process can be modelled as belief MDP 1H otherwnse an 709 my me Nol 3 Group","previews":{"preview":{"ref":"45210525-a5a8-4bab-a0a7-9451dfe8561e","$":"https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/e3f0f740-6fc8-415b-bba3-6a248bc21de2/track.jpg"}}},{"index":89,"time":5099640,"duration":9720,"relevance":-1,"hit":false,"text":"TU The optimal policies can be modelled as belief MDP QHCI cm I u l v H i Poupart P VIaSSIS N Hoey J Regan K An analytic solution to discrete BayeSIan reinforcement learning Proceedings of the 23rd international conference on Machine learning 2006 ACM 697 704 Nolzir glv Grove 51 m9 m9 at","previews":{"preview":{"ref":"45210525-a5a8-4bab-a0a7-9451dfe8561e","$":"https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/7020a465-0d40-4eb6-8454-6ecbcd819a40/track.jpg"}}},{"index":90,"time":5109360,"duration":20240,"relevance":-1,"hit":false,"text":"LU Applications QHCI I Clinical trials potential treatments for a disease to select from new patients or patient category at each round see c zcc oath n 3 3 zic annea r a t e e t a 5 is e 3 a at e at I Games Different moves at each round e g GO I Adaptive routing finding alternative paths also finding alternative roads for driving from A to B I Advertisement placemonts selection of an ad to display at the Webpage out of a finite set which can vary over time for each new Web page visitor Nolzr gthvouv 5 m9 049 as","previews":{"preview":{"ref":"45210525-a5a8-4bab-a0a7-9451dfe8561e","$":"https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/2f98a0d2-63ce-4bd0-bed8-56286b2bc93c/track.jpg"}}},{"index":91,"time":5129600,"duration":22960,"relevance":-1,"hit":false,"text":"TU Multi Armed Bandits and Clinical Trials cm hnp forIuneicom ZO1S 10 26 cancer clInicaI trial belmont repon Nolzr glvaouD s 709 04905","previews":{"preview":{"ref":"45210525-a5a8-4bab-a0a7-9451dfe8561e","$":"https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/b812d83e-361b-4812-82cb-e092240809af/track.jpg"}}},{"index":92,"time":5152560,"duration":41360,"relevance":-1,"hit":false,"text":"TU Example Clinical Trials in rare diseases QHCI cm Limitations of drug design for rare diseases due to I Lack of understanding of the underlying principles of the rare disease I Motivation Research advances I Unbalanced economic motivation cost benefit I Motivation Orphan Drug Act and other regulations I Unavailability of patients for standardtrials I This is the true bottleneck Villar S 5 Bowden J Wason J 2015 Mum armed Bandit Models for the Optimal Desngn of Clinical Trials Benefits and Challenges 199 215 don 10 1214 14 ST5504 Noizr glvaouD 54 m9 549 as","previews":{"preview":{"ref":"45210525-a5a8-4bab-a0a7-9451dfe8561e","$":"https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/bf4d1411-16c2-48b3-b9eb-4e720366858f/track.jpg"}}},{"index":93,"time":5193920,"duration":56640,"relevance":-1,"hit":false,"text":"TU Why should Bandit Strategies be of help here QHCI cm I The goal of Standard Randomized Controlled Trials RCT are a controlled learning setting I Control for Type and Type II errors dependent of trial size pm I In the case if the patient population N is smaller than the trial size HR T underpowered trial problem I If we change the goal to I learning sufficient to treat N as effectively as possible I then bandit strategies optimal policy for max the expected reward are perfectly suited 0 Kuleshov V Precup D 2014 Algorithms for mum armed bandit problems arXiv 1402 6028 Holznglmauv as m9 m9 at","previews":{"preview":{"ref":"45210525-a5a8-4bab-a0a7-9451dfe8561e","$":"https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/e3960148-e0b0-45c3-84ac-b8fc8bcd4ea1/track.jpg"}}},{"index":94,"time":5250560,"duration":8400,"relevance":-1,"hit":false,"text":"TU Learning vs Earning dilemma QHCI cm I Learning experimenting with all treatments I Earning selecting one treatment only based on experimentation results I Question 1 How much learning is best for an optimal treatment of N patients I Suppose N patients with a rare disease I Experimental Group E and control group C I e g control response rate pc and little information about experimental group I Question 2 How many allocations of treatment to E are necessary how much experimentation 709 L149 as News amp is","previews":{"preview":{"ref":"45210525-a5a8-4bab-a0a7-9451dfe8561e","$":"https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/7a161758-64bf-4e3e-97f9-825009822213/track.jpg"}}},{"index":95,"time":5258960,"duration":7760,"relevance":-1,"hit":false,"text":"TU Some Statistics on rare diseases orphan diseases QHCI cm I 7 000 different types more being discovered every day I 10 of the world population is suffering if all of the people with rare diseases lived in one country it would be the world s 3rd most populous country I 80 of rare diseases are genetic so are present throughout a person s lifetime even if symptoms do not immediately appear I 50 of the people affected by rare diseases are children I Are responsible for 35 of deaths in the first year of life I The prevalence distribution is skewed 80 of all rare disease patients are affected by 350 rare diseases I 50 of rare diseases do not hale a disease specific foundation supporting or researching their rare disease https g obaIgenesorg rare dIseases facts statistics https www honrch HONselect RareDIseases Nelmgu Gveuo as 709 049 as","previews":{"preview":{"ref":"45210525-a5a8-4bab-a0a7-9451dfe8561e","$":"https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/8451a69f-11ad-4902-a25a-d412b52cf810/track.jpg"}}},{"index":96,"time":5266720,"duration":20480,"relevance":-1,"hit":false,"text":"IU Example Rare Disease CADASIL 9H5 Cerebral Autosomal Dominant Arteriopathy with Subcortical Infarcts and Leukoencephalopathy is a hereditary disease affecting all the small cerebral arteries It causes subcortical infarcts and damages the white matter leukoencephalopathy and it is due to various mutations of the Notch3 gene Situated on chromosome 19 Joutel A et al 1996 thchB mutations ll l CADASIL a hereditary adult onset coriditioniausmg stroke and dementia Nature 383 6602 707 710 dOI 10 1038 383707a0 Nclzmglv Gvoup 59 m9 549 as","previews":{"preview":{"ref":"45210525-a5a8-4bab-a0a7-9451dfe8561e","$":"https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/bbb46923-b591-4330-ab8f-17a2e2d717c1/track.jpg"}}},{"index":97,"time":5287200,"duration":24360,"relevance":-1,"hit":false,"text":"TU CADASIL QHCI cm ChabrIaK H Joute A HDIchgans M Tournier Lasserve E Bousser M G 2009 CADASIL The Lancet Neurology 8 7 643 653 d0I hnp dx d0I org 10 1016 51474 4422 09 70127 9 ch new we 0 qu","previews":{"preview":{"ref":"45210525-a5a8-4bab-a0a7-9451dfe8561e","$":"https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/a1838e06-c47e-42bc-a268-38734a107ef3/track.jpg"}}},{"index":98,"time":5311560,"duration":5800,"relevance":-1,"hit":false,"text":"QHCI Conclusion and Future Challenges","previews":{"preview":{"ref":"45210525-a5a8-4bab-a0a7-9451dfe8561e","$":"https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/a521675a-8f35-40a8-acd3-8fe6991f0a3f/track.jpg"}}},{"index":99,"time":5317360,"duration":81280,"relevance":-1,"hit":false,"text":"TU Final conclusion QHCI cm Bandit strategy ls experimentation worth it for a small number N I Reconcile clinical trials and clinical practice Extensions should deal with randomization delayed responses and uncertainty around N Bayesian bandits need Online ML Bandits are a great source of inspirations and building blocks for solving many problems Future work convex optimization contextual combinatorial 0 Berry D A Fristedt BV 1985 Bandit problems sequential allocation of experiments Monographs on statistics and applied probability Springer News Group 93 m9 549 as","previews":{"preview":{"ref":"45210525-a5a8-4bab-a0a7-9451dfe8561e","$":"https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/71f34059-5140-44d2-bfc5-81624434cbd0/track.jpg"}}},{"index":100,"time":5398640,"duration":16347,"relevance":-1,"hit":false,"text":"V 7 m 9H ADA I 1 x v f I s Thankyou Nmmgu Evans 94 ms 549 as","previews":{"preview":{"ref":"45210525-a5a8-4bab-a0a7-9451dfe8561e","$":"https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/8e98bc46-c08d-4e0a-a37b-f8b7e96199f1/track.jpg"}}}]}}}}</article></body></html>

{"search-results":{"offset":0,"limit":100,"total":1,"searchTime":7,"query":"(id:e85fac19\\-c0af\\-4517\\-88ed\\-f961797258d8) AND oc_organization:mh_default_org AND (oc_acl_read:ROLE_ANONYMOUS) AND -oc_mediatype:Series AND -oc_deleted:[* TO *]","result":{"id":"e85fac19-c0af-4517-88ed-f961797258d8","org":"mh_default_org","mediapackage":{"duration":5414987,"id":"e85fac19-c0af-4517-88ed-f961797258d8","start":"2016-11-30T10:18:00Z","title":"06 Probabilisitc Graphical Models Part 2: From Bayesian Networks to Graph Bandits","series":"652fa592-2a86-43cb-8a9c-1c439dd00854","seriestitle":"[709049] Medical Informatics","creators":{"creator":"Holzinger A"},"license":"Creative Commons 3.0: Attribution-NonCommercial-NoDerivs","media":{"track":[{"id":"063565b9-8e04-432e-afdc-6f78f2b96faa","type":"presentation/delivery","ref":"track:ffd0aa9a-268d-4895-9d85-55a37afe0fa5","mimetype":"video/mp4","tags":{"tag":["engage-download","engage-streaming"]},"url":"https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/7b4367f8-ded8-42dd-8367-5b5385b8021f/track.mp4","checksum":{"type":"md5","$":"d138a82adf658c4de2d7dbfcd4ce4007"},"duration":5414987,"audio":{"id":"audio-1","device":"","encoder":{"type":"AAC (Advanced Audio Coding)"},"channels":2,"samplingrate":44100,"bitrate":64000},"video":{"id":"video-1","device":"","encoder":{"type":"H.264 / AVC / MPEG-4 AVC / MPEG-4 part 10"},"bitrate":202125,"framerate":25,"resolution":"1280x800"},"live":false},{"id":"45210525-a5a8-4bab-a0a7-9451dfe8561e","type":"presentation/delivery","ref":"track:ffd0aa9a-268d-4895-9d85-55a37afe0fa5","transport":"RTMP","mimetype":"video/mp4","tags":{"tag":["engage-download","engage-streaming"]},"url":"rtmp://wowza.tugraz.at/matterhorn_engage/mp4:mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/7b4367f8-ded8-42dd-8367-5b5385b8021f/track","checksum":{"type":"md5","$":"d138a82adf658c4de2d7dbfcd4ce4007"},"duration":5414987,"audio":{"id":"audio-1","device":"","encoder":{"type":"AAC (Advanced Audio Coding)"},"channels":2,"samplingrate":44100,"bitrate":64000},"video":{"id":"video-1","device":"","encoder":{"type":"H.264 / AVC / MPEG-4 AVC / MPEG-4 part 10"},"bitrate":202125,"framerate":25,"resolution":"1280x800"},"live":false}]},"metadata":{"catalog":[{"id":"cd56c60b-24f4-435d-adb8-92302f589a17","type":"dublincore/series","mimetype":"text/xml","tags":{"tag":["archive","engage-download"]},"url":"https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/70a3ffa5-3225-410b-9c34-f2fa4decb6c7/70a3ffa5_3225_410b_9c34_f2fa4decb6c7.xml","checksum":{"type":"md5","$":"4caaf586bd93d89f50877dc9fa14b3dd"}},{"id":"b6c0881d-f91f-490d-b811-5014abfaf960","type":"dublincore/episode","mimetype":"text/xml","tags":{"tag":["archive","engage-download"]},"url":"https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/4e36ced5-957c-48ea-8630-7b849984ca65/4e36ced5_957c_48ea_8630_7b849984ca65.xml","checksum":{"type":"md5","$":"5ee09835528d7136a96a56ed644c1e2a"}},{"id":"94c5ffb5-dd02-4494-b309-a71abde8ef14","type":"mpeg-7/text","ref":"track:ffd0aa9a-268d-4895-9d85-55a37afe0fa5","mimetype":"text/xml","tags":{"tag":"engage-download"},"url":"https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/9183f8ea-050f-41b3-9ba4-908e2d23a78e/slidetext.xml"},{"id":"4bd8ad77-b8f5-416a-895f-18dd34257b7a","type":"mpeg-7/text","ref":"track:track-1","mimetype":"text/xml","tags":{"tag":["archive","engage-download"]},"url":"https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/395f6144-4ec0-4ba4-ac5f-3df8215b07a2/catalog.xml","checksum":{"type":"md5","$":"f1597daed7f0f4d4bbd92da9c5634a16"}}]},"attachments":{"attachment":[{"id":"973e1493-1639-4a15-bbac-7b5699a858d1","type":"security/xacml","mimetype":"text/xml","tags":{"tag":"archive"},"url":"https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/security-policy-series/security-policy-series.xml","checksum":{"type":"md5","$":"4fe341748523f9c7035c9af3ecb78579"}},{"id":"f0b2f9e4-f4fc-4702-987d-d912e7a17f8b","type":"security/xacml+series","mimetype":"text/xml","tags":"","url":"https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/dd5b3ad7-c339-46ec-994d-40a3b9aa59d0/xacml.xml","size":0},{"id":"d6152ef5-0a06-4853-9368-85cd77d36a44","type":"presentation/segment+preview","ref":"track:45210525-a5a8-4bab-a0a7-9451dfe8561e;time=T00:21:58:560F1000","mimetype":"image/jpeg","tags":{"tag":"engage-download"},"url":"https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/1336c585-1504-49a0-a7bf-0fc9572f8c67/track.jpg","size":0},{"id":"3f16fecc-bea3-45b6-b610-213597caead5","type":"presentation/segment+preview","ref":"track:45210525-a5a8-4bab-a0a7-9451dfe8561e;time=T00:24:52:520F1000","mimetype":"image/jpeg","tags":{"tag":"engage-download"},"url":"https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/f00a235f-714a-419a-b93b-f9c733c277ee/track.jpg","size":0},{"id":"863b769a-b7a6-4904-bc51-c23ef55fdb00","type":"presentation/segment+preview","ref":"track:45210525-a5a8-4bab-a0a7-9451dfe8561e;time=T00:29:54:920F1000","mimetype":"image/jpeg","tags":{"tag":"engage-download"},"url":"https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/c579ee9b-7e66-4ac9-931d-55243d9236bc/track.jpg","size":0},{"id":"0d493e38-7153-4332-bc9d-fef87b80affa","type":"presentation/segment+preview","ref":"track:45210525-a5a8-4bab-a0a7-9451dfe8561e;time=T01:06:51:840F1000","mimetype":"image/jpeg","tags":{"tag":"engage-download"},"url":"https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/68c1787c-db47-4066-80ee-185a2205586c/track.jpg","size":0},{"id":"cb4856d6-94d6-4287-9831-c3f7de57fc42","type":"presentation/segment+preview","ref":"track:45210525-a5a8-4bab-a0a7-9451dfe8561e;time=T00:39:31:280F1000","mimetype":"image/jpeg","tags":{"tag":"engage-download"},"url":"https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/9ccda672-3ca8-40f7-a797-86c4ccc1daf6/track.jpg","size":0},{"id":"9480c549-73b0-4eb8-b108-a436230a608d","type":"presentation/segment+preview","ref":"track:45210525-a5a8-4bab-a0a7-9451dfe8561e;time=T00:49:59:120F1000","mimetype":"image/jpeg","tags":{"tag":"engage-download"},"url":"https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/c342a425-cabb-40a0-a732-535300542d22/track.jpg","size":0},{"id":"c0d5715d-1f0f-45c2-9fd6-b37faa14ed26","type":"presentation/segment+preview","ref":"track:45210525-a5a8-4bab-a0a7-9451dfe8561e;time=T00:17:14:839F1000","mimetype":"image/jpeg","tags":{"tag":"engage-download"},"url":"https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/540877d1-ab40-486f-a793-75be2b000005/track.jpg","size":0},{"id":"3ee01ed2-2025-4a76-9623-791bac9479c2","type":"presentation/segment+preview","ref":"track:45210525-a5a8-4bab-a0a7-9451dfe8561e;time=T01:28:37:360F1000","mimetype":"image/jpeg","tags":{"tag":"engage-download"},"url":"https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/71f34059-5140-44d2-bfc5-81624434cbd0/track.jpg","size":0},{"id":"fb6623e1-d80e-42c7-8cfa-43adc4e0ddc1","type":"presentation/segment+preview","ref":"track:45210525-a5a8-4bab-a0a7-9451dfe8561e;time=T00:27:43:320F1000","mimetype":"image/jpeg","tags":{"tag":"engage-download"},"url":"https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/82285407-758b-480f-ae22-6a0e64912c2a/track.jpg","size":0},{"id":"d0fc69bb-f012-4dd6-8bb0-de952e6a6290","type":"presentation/segment+preview","ref":"track:45210525-a5a8-4bab-a0a7-9451dfe8561e;time=T00:16:40:960F1000","mimetype":"image/jpeg","tags":{"tag":"engage-download"},"url":"https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/db18f079-68ae-447e-8e07-892c6773e8a2/track.jpg","size":0},{"id":"80ed5945-c89a-4382-9a3c-b30b5b41f26d","type":"presentation/segment+preview","ref":"track:45210525-a5a8-4bab-a0a7-9451dfe8561e;time=T00:14:12:360F1000","mimetype":"image/jpeg","tags":{"tag":"engage-download"},"url":"https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/4be1e541-ff90-44b7-8f84-c116cccc87b1/track.jpg","size":0},{"id":"e80c368c-2240-4ec7-8ac8-278724c59962","type":"presentation/segment+preview","ref":"track:45210525-a5a8-4bab-a0a7-9451dfe8561e;time=T01:05:08:880F1000","mimetype":"image/jpeg","tags":{"tag":"engage-download"},"url":"https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/835ca2af-7235-4911-aa10-c3442fcbf6a6/track.jpg","size":0},{"id":"25024805-77f0-437a-9045-d4fdc08181f6","type":"presentation/segment+preview","ref":"track:45210525-a5a8-4bab-a0a7-9451dfe8561e;time=T00:12:20:280F1000","mimetype":"image/jpeg","tags":{"tag":"engage-download"},"url":"https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/15d297a9-96b7-41dd-894d-6e4fa92d1a84/track.jpg","size":0},{"id":"ff43a14f-c068-41ad-ba53-5b94a4758a58","type":"presentation/segment+preview","ref":"track:45210525-a5a8-4bab-a0a7-9451dfe8561e;time=T01:07:23:680F1000","mimetype":"image/jpeg","tags":{"tag":"engage-download"},"url":"https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/1e052c0a-5870-4dbe-ad1b-3c15715ccbd2/track.jpg","size":0},{"id":"b50bdcd3-9fef-425f-ba28-d03488e5c01c","type":"presentation/segment+preview","ref":"track:45210525-a5a8-4bab-a0a7-9451dfe8561e;time=T00:43:10:920F1000","mimetype":"image/jpeg","tags":{"tag":"engage-download"},"url":"https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/e9325995-b460-4c65-9c8a-72f0e3972b25/track.jpg","size":0},{"id":"6272e0b8-5f77-43c2-a2ae-85c64813deff","type":"presentation/segment+preview","ref":"track:45210525-a5a8-4bab-a0a7-9451dfe8561e;time=T01:26:33:920F1000","mimetype":"image/jpeg","tags":{"tag":"engage-download"},"url":"https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/e3960148-e0b0-45c3-84ac-b8fc8bcd4ea1/track.jpg","size":0},{"id":"b0883db4-e094-4689-b42c-9a3e6ddd4d43","type":"presentation/segment+preview","ref":"track:45210525-a5a8-4bab-a0a7-9451dfe8561e;time=T00:20:56:40F1000","mimetype":"image/jpeg","tags":{"tag":"engage-download"},"url":"https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/a4fd407c-ab7b-4a5e-8434-7dcb1100fd5e/track.jpg","size":0},{"id":"f6ddd8d0-9557-421f-8fa2-0a11d3d51eab","type":"presentation/segment+preview","ref":"track:45210525-a5a8-4bab-a0a7-9451dfe8561e;time=T01:24:53:80F1000","mimetype":"image/jpeg","tags":{"tag":"engage-download"},"url":"https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/e3f0f740-6fc8-415b-bba3-6a248bc21de2/track.jpg","size":0},{"id":"36d183af-6fe9-4a69-9d51-e43dd53325f2","type":"presentation/segment+preview","ref":"track:45210525-a5a8-4bab-a0a7-9451dfe8561e;time=T00:52:04:880F1000","mimetype":"image/jpeg","tags":{"tag":"engage-download"},"url":"https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/d6759621-7816-423d-9969-dd4cccb1b4ba/track.jpg","size":0},{"id":"4ed36e18-5feb-495c-8c33-7e2ebe094952","type":"presentation/segment+preview","ref":"track:45210525-a5a8-4bab-a0a7-9451dfe8561e;time=T00:12:13:800F1000","mimetype":"image/jpeg","tags":{"tag":"engage-download"},"url":"https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/15e202d1-27b0-496e-ac89-ca801fca8e2f/track.jpg","size":0},{"id":"fee30829-93c6-4174-ab7e-aeb51218eec1","type":"presentation/segment+preview","ref":"track:45210525-a5a8-4bab-a0a7-9451dfe8561e;time=T00:24:37:760F1000","mimetype":"image/jpeg","tags":{"tag":"engage-download"},"url":"https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/6caf36e0-10ef-496a-adc7-2c4bc89ef2bd/track.jpg","size":0},{"id":"3ce34bd0-a5c1-4765-9141-083f32b5ba4d","type":"presentation/segment+preview","ref":"track:45210525-a5a8-4bab-a0a7-9451dfe8561e;time=T01:02:37:960F1000","mimetype":"image/jpeg","tags":{"tag":"engage-download"},"url":"https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/17e75605-f22c-49d6-9fbc-c0f4900bcfbc/track.jpg","size":0},{"id":"c7da78b9-63ef-47b4-ba14-05ae686ad8df","type":"presentation/segment+preview","ref":"track:45210525-a5a8-4bab-a0a7-9451dfe8561e;time=T01:04:28:600F1000","mimetype":"image/jpeg","tags":{"tag":"engage-download"},"url":"https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/f0de732e-5ddd-4caf-ab64-ba2e7939cb70/track.jpg","size":0},{"id":"a78a5636-c0d8-48fc-8890-d30851e28933","type":"presentation/segment+preview","ref":"track:45210525-a5a8-4bab-a0a7-9451dfe8561e;time=T01:25:52:560F1000","mimetype":"image/jpeg","tags":{"tag":"engage-download"},"url":"https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/bf4d1411-16c2-48b3-b9eb-4e720366858f/track.jpg","size":0},{"id":"7de914dd-543d-4840-9295-054a79829d9a","type":"presentation/segment+preview","ref":"track:45210525-a5a8-4bab-a0a7-9451dfe8561e;time=T00:28:40:360F1000","mimetype":"image/jpeg","tags":{"tag":"engage-download"},"url":"https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/6446e2b3-f7a8-4e64-9df1-46d63281b45b/track.jpg","size":0},{"id":"50f9aad9-c53e-4095-844a-f82781db8d58","type":"presentation/segment+preview","ref":"track:45210525-a5a8-4bab-a0a7-9451dfe8561e;time=T01:14:00:600F1000","mimetype":"image/jpeg","tags":{"tag":"engage-download"},"url":"https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/e2acd983-4a95-424d-8ad9-73111a1d8178/track.jpg","size":0},{"id":"7defadb9-2ca1-48b0-883e-57995fa247fa","type":"presentation/segment+preview","ref":"track:45210525-a5a8-4bab-a0a7-9451dfe8561e;time=T00:32:12:280F1000","mimetype":"image/jpeg","tags":{"tag":"engage-download"},"url":"https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/b29a2412-d7de-4f97-8434-0025dcb4a7a3/track.jpg","size":0},{"id":"ad14abad-e5cc-4d90-8fdc-fc46968ebaf7","type":"presentation/segment+preview","ref":"track:45210525-a5a8-4bab-a0a7-9451dfe8561e;time=T01:20:21:40F1000","mimetype":"image/jpeg","tags":{"tag":"engage-download"},"url":"https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/98f14e9e-0bd5-4814-accc-27dd0c253161/track.jpg","size":0},{"id":"097e20ed-aa55-4a8b-a26e-c3c7db3f1921","type":"presentation/segment+preview","ref":"track:45210525-a5a8-4bab-a0a7-9451dfe8561e;time=T00:45:48:400F1000","mimetype":"image/jpeg","tags":{"tag":"engage-download"},"url":"https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/6ec9c41f-f1dd-41d8-966c-cc852d005075/track.jpg","size":0},{"id":"3c86b5ca-4260-4b52-a07a-864bd48965b0","type":"presentation/segment+preview","ref":"track:45210525-a5a8-4bab-a0a7-9451dfe8561e;time=T00:01:54:560F1000","mimetype":"image/jpeg","tags":{"tag":"engage-download"},"url":"https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/274b400c-9e92-4b98-8b8a-52c3b852ad03/track.jpg","size":0},{"id":"3b56ec9d-62fa-46ec-abe5-70becc1bbe92","type":"presentation/segment+preview","ref":"track:45210525-a5a8-4bab-a0a7-9451dfe8561e;time=T00:08:17:360F1000","mimetype":"image/jpeg","tags":{"tag":"engage-download"},"url":"https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/6e951bc3-5e85-4aa8-a0b2-5238edd24b13/track.jpg","size":0},{"id":"5a477fec-238c-4117-82f0-18ce3dba34d4","type":"presentation/segment+preview","ref":"track:45210525-a5a8-4bab-a0a7-9451dfe8561e;time=T01:03:05:600F1000","mimetype":"image/jpeg","tags":{"tag":"engage-download"},"url":"https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/82b5188e-6bfa-463f-a815-e6bfc091c6eb/track.jpg","size":0},{"id":"f8ec2f9f-3bba-4f61-b56f-fb740a7427cb","type":"presentation/segment+preview","ref":"track:45210525-a5a8-4bab-a0a7-9451dfe8561e;time=T00:09:43:480F1000","mimetype":"image/jpeg","tags":{"tag":"engage-download"},"url":"https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/e8c5b1bf-d1de-4806-b6ac-a26482b5deac/track.jpg","size":0},{"id":"db089d92-a64b-4dd0-9b97-42bc5d8bc434","type":"presentation/segment+preview","ref":"track:45210525-a5a8-4bab-a0a7-9451dfe8561e;time=T00:33:59:720F1000","mimetype":"image/jpeg","tags":{"tag":"engage-download"},"url":"https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/39335258-3766-42e2-b74b-f21bfc4b5a12/track.jpg","size":0},{"id":"8689b77a-2344-461e-8032-d6d3cc58a2df","type":"presentation/segment+preview","ref":"track:45210525-a5a8-4bab-a0a7-9451dfe8561e;time=T00:34:36:40F1000","mimetype":"image/jpeg","tags":{"tag":"engage-download"},"url":"https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/be112ef5-6009-4bc3-a476-4ecb3a9d77a0/track.jpg","size":0},{"id":"4e9e957f-0d56-4861-bde9-2dc207ed2fdf","type":"presentation/segment+preview","ref":"track:45210525-a5a8-4bab-a0a7-9451dfe8561e;time=T00:19:19:480F1000","mimetype":"image/jpeg","tags":{"tag":"engage-download"},"url":"https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/a80a39c5-c445-4304-882f-6887423f6e96/track.jpg","size":0},{"id":"4f7d419a-1af0-4d24-9af2-b6e528c8facc","type":"presentation/segment+preview","ref":"track:45210525-a5a8-4bab-a0a7-9451dfe8561e;time=T00:09:16:200F1000","mimetype":"image/jpeg","tags":{"tag":"engage-download"},"url":"https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/3415228d-5667-4ed7-8379-69d95befa033/track.jpg","size":0},{"id":"15fc84d6-d6ad-4e0f-aa06-33c79a9ce129","type":"presentation/segment+preview","ref":"track:45210525-a5a8-4bab-a0a7-9451dfe8561e;time=T01:04:55:280F1000","mimetype":"image/jpeg","tags":{"tag":"engage-download"},"url":"https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/ca1ad350-3192-41f2-b2be-734b1bba7373/track.jpg","size":0},{"id":"b1ba5ad4-88fb-4ec1-82eb-3f74e5d9b39f","type":"presentation/segment+preview","ref":"track:45210525-a5a8-4bab-a0a7-9451dfe8561e;time=T01:24:59:640F1000","mimetype":"image/jpeg","tags":{"tag":"engage-download"},"url":"https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/7020a465-0d40-4eb6-8454-6ecbcd819a40/track.jpg","size":0},{"id":"bb5a0700-1b5b-4bdb-b06a-842ea9355d10","type":"presentation/segment+preview","ref":"track:45210525-a5a8-4bab-a0a7-9451dfe8561e;time=T00:49:03:680F1000","mimetype":"image/jpeg","tags":{"tag":"engage-download"},"url":"https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/6dfc7169-200b-4084-b780-bfd15c6e3651/track.jpg","size":0},{"id":"3f235ee8-fd35-4678-ad0c-90faaeb3b88a","type":"presentation/segment+preview","ref":"track:45210525-a5a8-4bab-a0a7-9451dfe8561e;time=T01:28:07:200F1000","mimetype":"image/jpeg","tags":{"tag":"engage-download"},"url":"https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/a1838e06-c47e-42bc-a268-38734a107ef3/track.jpg","size":0},{"id":"cc4ede90-3c22-4a0b-ba64-129cb94c97e7","type":"presentation/segment+preview","ref":"track:45210525-a5a8-4bab-a0a7-9451dfe8561e;time=T00:56:34:40F1000","mimetype":"image/jpeg","tags":{"tag":"engage-download"},"url":"https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/eee918ba-be51-4bd6-97a0-ab7f740bd4c7/track.jpg","size":0},{"id":"b045c314-1484-457d-bdf9-db372e84e80a","type":"presentation/segment+preview","ref":"track:45210525-a5a8-4bab-a0a7-9451dfe8561e;time=T00:00:00:0F1000","mimetype":"image/jpeg","tags":{"tag":"engage-download"},"url":"https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/d2128d70-db12-49ec-99e4-3f00f21cc51f/track.jpg","size":0},{"id":"f54e9273-e13d-45e7-98e3-babbc856ebc5","type":"presentation/segment+preview","ref":"track:45210525-a5a8-4bab-a0a7-9451dfe8561e;time=T00:48:07:520F1000","mimetype":"image/jpeg","tags":{"tag":"engage-download"},"url":"https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/dffc5f4a-3b7d-4af9-9753-04151aaca684/track.jpg","size":0},{"id":"efcc30fd-9f83-43cd-8ffe-d72462ce8faa","type":"presentation/segment+preview","ref":"track:45210525-a5a8-4bab-a0a7-9451dfe8561e;time=T00:09:32:920F1000","mimetype":"image/jpeg","tags":{"tag":"engage-download"},"url":"https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/5891df7b-cac0-4cf9-b64f-d1049234811a/track.jpg","size":0},{"id":"f3c428be-b543-4e25-a4b5-7edd15831998","type":"presentation/segment+preview","ref":"track:45210525-a5a8-4bab-a0a7-9451dfe8561e;time=T00:53:32:200F1000","mimetype":"image/jpeg","tags":{"tag":"engage-download"},"url":"https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/8c57954f-4451-45f0-98ba-131bb2f55272/track.jpg","size":0},{"id":"8aba5bc7-3ed6-4e84-aee1-7608e1dde532","type":"presentation/segment+preview","ref":"track:45210525-a5a8-4bab-a0a7-9451dfe8561e;time=T01:24:28:880F1000","mimetype":"image/jpeg","tags":{"tag":"engage-download"},"url":"https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/628b2a11-514b-4fa3-8ba9-9aca78ad4e2e/track.jpg","size":0},{"id":"7538d46d-7538-4ecf-ac09-1ab6ecb1d9e6","type":"presenter/player+preview","mimetype":"image/png","tags":{"tag":["archive","engage-download"]},"url":"https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/e7e94061-3cfb-452e-a1c4-e6a24dd52fc5/coverimage.png","size":0},{"id":"bce001ac-79e8-449d-aa6a-b1e1785df3e1","type":"presentation/segment+preview","ref":"track:45210525-a5a8-4bab-a0a7-9451dfe8561e;time=T01:05:48:880F1000","mimetype":"image/jpeg","tags":{"tag":"engage-download"},"url":"https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/9c3c7f8a-c597-4d07-bc3b-62bae8b79223/track.jpg","size":0},{"id":"44aad8d9-6ee0-4422-a6c3-070ea404ce0d","type":"presentation/segment+preview","ref":"track:45210525-a5a8-4bab-a0a7-9451dfe8561e;time=T01:18:14:240F1000","mimetype":"image/jpeg","tags":{"tag":"engage-download"},"url":"https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/3016dfa2-6f63-4838-905a-56503c1ddea3/track.jpg","size":0},{"id":"818531d9-0657-401d-a8e0-921fa01d5639","type":"presentation/segment+preview","ref":"track:45210525-a5a8-4bab-a0a7-9451dfe8561e;time=T01:21:19:240F1000","mimetype":"image/jpeg","tags":{"tag":"engage-download"},"url":"https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/2d278a9a-1a23-449f-8cdd-718ef36b6bf5/track.jpg","size":0},{"id":"f7652451-f6fd-4a68-8d00-dada76109346","type":"presentation/segment+preview","ref":"track:45210525-a5a8-4bab-a0a7-9451dfe8561e;time=T01:12:05:0F1000","mimetype":"image/jpeg","tags":{"tag":"engage-download"},"url":"https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/d7ec3d3e-0242-4bb6-ae8e-10e8cc9d8b47/track.jpg","size":0},{"id":"6983e8ae-51b4-41e0-981a-9e6ea7ab5179","type":"presentation/segment+preview","ref":"track:45210525-a5a8-4bab-a0a7-9451dfe8561e;time=T01:01:26:560F1000","mimetype":"image/jpeg","tags":{"tag":"engage-download"},"url":"https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/661b3521-6399-4131-b8d2-bfd5c1d47441/track.jpg","size":0},{"id":"6027a2c7-db2e-4005-9645-a8a5e65cd9f6","type":"presentation/segment+preview","ref":"track:45210525-a5a8-4bab-a0a7-9451dfe8561e;time=T01:13:20:400F1000","mimetype":"image/jpeg","tags":{"tag":"engage-download"},"url":"https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/63a39edf-fc5e-43b7-8b22-dad80bdda431/track.jpg","size":0},{"id":"12f87a68-a57b-4949-a25a-4640fc043669","type":"presentation/segment+preview","ref":"track:45210525-a5a8-4bab-a0a7-9451dfe8561e;time=T00:20:14:240F1000","mimetype":"image/jpeg","tags":{"tag":"engage-download"},"url":"https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/f22280d2-d58d-4f7b-a45d-b93526d70fbe/track.jpg","size":0},{"id":"5add9872-dd04-433b-ba69-84506a3bc438","type":"presentation/segment+preview","ref":"track:45210525-a5a8-4bab-a0a7-9451dfe8561e;time=T00:54:42:680F1000","mimetype":"image/jpeg","tags":{"tag":"engage-download"},"url":"https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/5f67c6bd-1941-4052-a696-96aa17daefbc/track.jpg","size":0},{"id":"5ef67b12-360e-41d3-955e-2531c918e187","type":"presentation/segment+preview","ref":"track:45210525-a5a8-4bab-a0a7-9451dfe8561e;time=T00:54:51:960F1000","mimetype":"image/jpeg","tags":{"tag":"engage-download"},"url":"https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/35a92b3f-2e1e-4308-9c90-0a2b39f6bb51/track.jpg","size":0},{"id":"72b38d5e-8f37-4319-91fc-5807e3662ab4","type":"presentation/segment+preview","ref":"track:45210525-a5a8-4bab-a0a7-9451dfe8561e;time=T00:38:53:920F1000","mimetype":"image/jpeg","tags":{"tag":"engage-download"},"url":"https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/9af7c54f-3da7-49c3-852e-bf3b66f6aded/track.jpg","size":0},{"id":"97a40235-0821-4232-8643-655c868636d0","type":"presentation/segment+preview","ref":"track:45210525-a5a8-4bab-a0a7-9451dfe8561e;time=T00:41:14:320F1000","mimetype":"image/jpeg","tags":{"tag":"engage-download"},"url":"https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/a0c59d71-959f-4c49-ab51-5e462bf4c221/track.jpg","size":0},{"id":"92744892-71fc-4553-acc5-944b2411df4e","type":"presentation/segment+preview","ref":"track:45210525-a5a8-4bab-a0a7-9451dfe8561e;time=T01:06:09:360F1000","mimetype":"image/jpeg","tags":{"tag":"engage-download"},"url":"https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/33c4b33a-8a0a-45f8-9086-354b82a36b26/track.jpg","size":0},{"id":"04da209b-bf32-4bb9-b1e6-410eed901c4f","type":"presentation/segment+preview","ref":"track:45210525-a5a8-4bab-a0a7-9451dfe8561e;time=T01:27:30:560F1000","mimetype":"image/jpeg","tags":{"tag":"engage-download"},"url":"https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/7a161758-64bf-4e3e-97f9-825009822213/track.jpg","size":0},{"id":"e2de28e8-9bac-4c38-9ede-e20c4843dbf2","type":"presentation/segment+preview","ref":"track:45210525-a5a8-4bab-a0a7-9451dfe8561e;time=T00:42:31:360F1000","mimetype":"image/jpeg","tags":{"tag":"engage-download"},"url":"https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/f0952a7a-d650-4656-90f4-bb7e7dc5e949/track.jpg","size":0},{"id":"d05d8cf6-e4d7-4374-b1b3-2651cde1d98e","type":"presentation/segment+preview","ref":"track:45210525-a5a8-4bab-a0a7-9451dfe8561e;time=T01:09:20:799F1000","mimetype":"image/jpeg","tags":{"tag":"engage-download"},"url":"https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/4cef5745-e19a-4571-9b11-84b3948d8d8e/track.jpg","size":0},{"id":"b84c1dab-06fc-4f11-8c85-3335fc042429","type":"presentation/segment+preview","ref":"track:45210525-a5a8-4bab-a0a7-9451dfe8561e;time=T00:33:10:880F1000","mimetype":"image/jpeg","tags":{"tag":"engage-download"},"url":"https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/5b258709-ba4f-46e0-aae6-49a61a64e25b/track.jpg","size":0},{"id":"d841ccdc-4336-405a-ac19-0a381c327709","type":"presentation/segment+preview","ref":"track:45210525-a5a8-4bab-a0a7-9451dfe8561e;time=T00:26:44:960F1000","mimetype":"image/jpeg","tags":{"tag":"engage-download"},"url":"https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/519507a6-eec8-4782-b57d-9d09f351975e/track.jpg","size":0},{"id":"19acf831-1ad7-40d4-ab04-b5caa5931973","type":"presentation/segment+preview","ref":"track:45210525-a5a8-4bab-a0a7-9451dfe8561e;time=T00:31:00:480F1000","mimetype":"image/jpeg","tags":{"tag":"engage-download"},"url":"https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/83fb9e15-4f84-48c0-8ef9-0ac99e2300cc/track.jpg","size":0},{"id":"ff5278dd-824c-4e92-beaf-7e57ffcc5c90","type":"presentation/segment+preview","ref":"track:45210525-a5a8-4bab-a0a7-9451dfe8561e;time=T01:25:09:360F1000","mimetype":"image/jpeg","tags":{"tag":"engage-download"},"url":"https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/2f98a0d2-63ce-4bd0-bed8-56286b2bc93c/track.jpg","size":0},{"id":"f608aa9b-73ca-4986-8d14-b970b4696f44","type":"presentation/segment+preview","ref":"track:45210525-a5a8-4bab-a0a7-9451dfe8561e;time=T00:51:45:480F1000","mimetype":"image/jpeg","tags":{"tag":"engage-download"},"url":"https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/53f08c19-ebc5-4ce7-9ddb-2681eb42c00c/track.jpg","size":0},{"id":"d3745f34-1d3f-4738-914a-68ef9f1b6901","type":"presentation/segment+preview","ref":"track:45210525-a5a8-4bab-a0a7-9451dfe8561e;time=T01:04:37:400F1000","mimetype":"image/jpeg","tags":{"tag":"engage-download"},"url":"https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/addb0b35-fee3-4ef9-9680-e9d913c71982/track.jpg","size":0},{"id":"3480a20a-c8c4-4eee-9a63-73a75a2faa73","type":"presentation/segment+preview","ref":"track:45210525-a5a8-4bab-a0a7-9451dfe8561e;time=T01:27:38:960F1000","mimetype":"image/jpeg","tags":{"tag":"engage-download"},"url":"https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/8451a69f-11ad-4902-a25a-d412b52cf810/track.jpg","size":0},{"id":"d4b2a422-ffca-4690-80d6-84dbd79e33fe","type":"presentation/segment+preview","ref":"track:45210525-a5a8-4bab-a0a7-9451dfe8561e;time=T01:29:58:640F1000","mimetype":"image/jpeg","tags":{"tag":"engage-download"},"url":"https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/8e98bc46-c08d-4e0a-a37b-f8b7e96199f1/track.jpg","size":0},{"id":"346c06cc-569b-47e3-b84d-b8b93aa8ab38","type":"presentation/segment+preview","ref":"track:45210525-a5a8-4bab-a0a7-9451dfe8561e;time=T00:46:28:320F1000","mimetype":"image/jpeg","tags":{"tag":"engage-download"},"url":"https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/ffac8015-7952-4061-a9c5-00c95b5de942/track.jpg","size":0},{"id":"5abe3c0c-6a1b-4ab0-8818-f6866bd82ab8","type":"presentation/segment+preview","ref":"track:45210525-a5a8-4bab-a0a7-9451dfe8561e;time=T00:43:16:200F1000","mimetype":"image/jpeg","tags":{"tag":"engage-download"},"url":"https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/e62c918e-485c-45c6-9e7f-742c7e1446c4/track.jpg","size":0},{"id":"13bdb656-6534-41bd-8528-4aefce7646aa","type":"presentation/segment+preview","ref":"track:45210525-a5a8-4bab-a0a7-9451dfe8561e;time=T01:08:07:920F1000","mimetype":"image/jpeg","tags":{"tag":"engage-download"},"url":"https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/f16d5787-f7b0-4635-8675-8ea5c56411c1/track.jpg","size":0},{"id":"714ded71-ac02-4b26-9369-05670137a278","type":"presentation/segment+preview","ref":"track:45210525-a5a8-4bab-a0a7-9451dfe8561e;time=T00:26:12:0F1000","mimetype":"image/jpeg","tags":{"tag":"engage-download"},"url":"https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/e376250c-5db2-4426-b393-bbc4e7a2d19a/track.jpg","size":0},{"id":"240efa5e-dbb8-4fcf-a933-8306e345a8a2","type":"presentation/segment+preview","ref":"track:45210525-a5a8-4bab-a0a7-9451dfe8561e;time=T00:05:53:160F1000","mimetype":"image/jpeg","tags":{"tag":"engage-download"},"url":"https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/d5e51fdc-4d7c-42ce-a3de-dae9ffdb8103/track.jpg","size":0},{"id":"db521ab6-3d3c-4980-8030-e155b7afb3a7","type":"presentation/segment+preview","ref":"track:45210525-a5a8-4bab-a0a7-9451dfe8561e;time=T00:23:42:160F1000","mimetype":"image/jpeg","tags":{"tag":"engage-download"},"url":"https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/a2295e0a-f7ef-41fd-847b-40679ae28d5d/track.jpg","size":0},{"id":"d1b9dbb3-b8bf-4772-a9c2-44367f60bb15","type":"presentation/segment+preview","ref":"track:45210525-a5a8-4bab-a0a7-9451dfe8561e;time=T00:35:52:0F1000","mimetype":"image/jpeg","tags":{"tag":"engage-download"},"url":"https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/f9ba87ff-9ee0-4adc-a383-1b00646b7527/track.jpg","size":0},{"id":"634d7de0-430a-497c-99ee-239a4aa3ff53","type":"presentation/segment+preview","ref":"track:45210525-a5a8-4bab-a0a7-9451dfe8561e;time=T00:36:44:280F1000","mimetype":"image/jpeg","tags":{"tag":"engage-download"},"url":"https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/91059181-d401-4617-b548-81e0723871da/track.jpg","size":0},{"id":"ba5ef2e4-86c6-4d27-9c76-627b329265f5","type":"presentation/segment+preview","ref":"track:45210525-a5a8-4bab-a0a7-9451dfe8561e;time=T01:08:43:120F1000","mimetype":"image/jpeg","tags":{"tag":"engage-download"},"url":"https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/13803432-0117-4425-9cca-21ef610c3271/track.jpg","size":0},{"id":"8df8e840-81ff-49a5-abb2-e2bd5a3f28c0","type":"presentation/segment+preview","ref":"track:45210525-a5a8-4bab-a0a7-9451dfe8561e;time=T01:07:15:320F1000","mimetype":"image/jpeg","tags":{"tag":"engage-download"},"url":"https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/9abd0d03-0891-458b-8006-0a911695502d/track.jpg","size":0},{"id":"7e35f66c-e6ff-4936-8be3-92c4b1e9de42","type":"presentation/segment+preview","ref":"track:45210525-a5a8-4bab-a0a7-9451dfe8561e;time=T00:35:08:320F1000","mimetype":"image/jpeg","tags":{"tag":"engage-download"},"url":"https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/f1acca10-7339-48f3-b8a8-45d235c6b0c9/track.jpg","size":0},{"id":"69ed39a7-86d8-40cf-9ae8-03292ae5a785","type":"presentation/segment+preview","ref":"track:45210525-a5a8-4bab-a0a7-9451dfe8561e;time=T00:57:31:920F1000","mimetype":"image/jpeg","tags":{"tag":"engage-download"},"url":"https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/57a64a0d-b0dc-4640-8237-a4df5ef3f941/track.jpg","size":0},{"id":"f0723b44-bc94-4997-b788-9150169dca73","type":"presentation/segment+preview","ref":"track:45210525-a5a8-4bab-a0a7-9451dfe8561e;time=T00:59:43:880F1000","mimetype":"image/jpeg","tags":{"tag":"engage-download"},"url":"https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/de1e857b-3297-4b8e-8197-4734ae81cf41/track.jpg","size":0},{"id":"55c49d99-e3df-40ca-a131-fb5df173440d","type":"presentation/segment+preview","ref":"track:45210525-a5a8-4bab-a0a7-9451dfe8561e;time=T01:28:31:560F1000","mimetype":"image/jpeg","tags":{"tag":"engage-download"},"url":"https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/a521675a-8f35-40a8-acd3-8fe6991f0a3f/track.jpg","size":0},{"id":"469ec150-b26c-4ca1-a151-af9d7cea92e1","type":"presentation/segment+preview","ref":"track:45210525-a5a8-4bab-a0a7-9451dfe8561e;time=T00:35:00:960F1000","mimetype":"image/jpeg","tags":{"tag":"engage-download"},"url":"https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/ea79344c-865a-4898-970a-5f1caead5f7a/track.jpg","size":0},{"id":"52cdd696-38e5-44c4-9d7a-e4f3a69738a3","type":"presentation/segment+preview","ref":"track:45210525-a5a8-4bab-a0a7-9451dfe8561e;time=T00:25:20:520F1000","mimetype":"image/jpeg","tags":{"tag":"engage-download"},"url":"https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/6d5a6667-c1ad-476b-9d18-56ec9de25081/track.jpg","size":0},{"id":"87dbf3ff-624a-443e-8af5-417551f26a51","type":"presentation/segment+preview","ref":"track:45210525-a5a8-4bab-a0a7-9451dfe8561e;time=T00:19:25:880F1000","mimetype":"image/jpeg","tags":{"tag":"engage-download"},"url":"https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/ef3d6ebe-737c-42a6-aedd-9fc8ee781d11/track.jpg","size":0},{"id":"a08de618-f4c2-4db4-8fb8-33456baf6298","type":"presentation/segment+preview","ref":"track:45210525-a5a8-4bab-a0a7-9451dfe8561e;time=T01:24:05:480F1000","mimetype":"image/jpeg","tags":{"tag":"engage-download"},"url":"https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/1b3fc3a1-9d79-42f4-9d9c-9809b2121efa/track.jpg","size":0},{"id":"dfa2848a-b93e-42ae-aac5-b3b39d65f4b6","type":"presentation/segment+preview","ref":"track:45210525-a5a8-4bab-a0a7-9451dfe8561e;time=T00:50:17:480F1000","mimetype":"image/jpeg","tags":{"tag":"engage-download"},"url":"https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/9724eb4c-9371-41a2-84b0-b216a65730fa/track.jpg","size":0},{"id":"f7bd3d95-f4c4-4779-9789-e7fc5f61cbdd","type":"presentation/segment+preview","ref":"track:45210525-a5a8-4bab-a0a7-9451dfe8561e;time=T01:12:50:680F1000","mimetype":"image/jpeg","tags":{"tag":"engage-download"},"url":"https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/cd276b03-f8c6-443b-b416-4d4ed7385e1b/track.jpg","size":0},{"id":"dcda8508-54f8-433b-9bfe-acddea561a27","type":"presentation/segment+preview","ref":"track:45210525-a5a8-4bab-a0a7-9451dfe8561e;time=T00:47:50:640F1000","mimetype":"image/jpeg","tags":{"tag":"engage-download"},"url":"https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/e966ff8e-ee9d-410a-bcb3-6f6f9d9de1c8/track.jpg","size":0},{"id":"2248f820-40ee-40b5-aad3-6431c963a921","type":"presentation/segment+preview","ref":"track:45210525-a5a8-4bab-a0a7-9451dfe8561e;time=T01:25:29:600F1000","mimetype":"image/jpeg","tags":{"tag":"engage-download"},"url":"https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/b812d83e-361b-4812-82cb-e092240809af/track.jpg","size":0},{"id":"570aa523-bc20-4b32-b423-903e3e149bc3","type":"presentation/segment+preview","ref":"track:45210525-a5a8-4bab-a0a7-9451dfe8561e;time=T00:12:37:160F1000","mimetype":"image/jpeg","tags":{"tag":"engage-download"},"url":"https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/3c970d74-059c-4d0a-af1d-feed5d3951f9/track.jpg","size":0},{"id":"8fad7d30-c681-47e4-bc0d-28e117fe58b0","type":"presentation/segment+preview","ref":"track:45210525-a5a8-4bab-a0a7-9451dfe8561e;time=T00:42:21:680F1000","mimetype":"image/jpeg","tags":{"tag":"engage-download"},"url":"https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/a1fe62b1-e6b6-4c49-9f7c-15131173bfac/track.jpg","size":0},{"id":"784fe4d6-c8c6-48ae-9860-adbedfafd945","type":"presentation/segment+preview","ref":"track:45210525-a5a8-4bab-a0a7-9451dfe8561e;time=T01:03:53:640F1000","mimetype":"image/jpeg","tags":{"tag":"engage-download"},"url":"https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/d5ecddad-75be-476a-9041-f408f91c7801/track.jpg","size":0},{"id":"0608259c-0d67-4022-bf57-6933b2fe870c","type":"presentation/segment+preview","ref":"track:45210525-a5a8-4bab-a0a7-9451dfe8561e;time=T00:00:42:800F1000","mimetype":"image/jpeg","tags":{"tag":"engage-download"},"url":"https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/76e6597e-d06c-462c-9370-55bbeb374af2/track.jpg","size":0},{"id":"623ba10e-6081-45d7-a8b8-74d4726f0221","type":"presentation/segment+preview","ref":"track:45210525-a5a8-4bab-a0a7-9451dfe8561e;time=T00:38:46:440F1000","mimetype":"image/jpeg","tags":{"tag":"engage-download"},"url":"https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/3166fdd2-7067-4d14-a973-178a413dd678/track.jpg","size":0},{"id":"4cc1f05f-97bc-4694-88f6-95706cf91e18","type":"presentation/segment+preview","ref":"track:45210525-a5a8-4bab-a0a7-9451dfe8561e;time=T00:40:50:440F1000","mimetype":"image/jpeg","tags":{"tag":"engage-download"},"url":"https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/72156a34-ffef-4961-9abf-8218c8d3f9f0/track.jpg","size":0},{"id":"882ae3b3-b461-4233-a9cb-09fd82856613","type":"presentation/segment+preview","ref":"track:45210525-a5a8-4bab-a0a7-9451dfe8561e;time=T01:27:46:720F1000","mimetype":"image/jpeg","tags":{"tag":"engage-download"},"url":"https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/bbb46923-b591-4330-ab8f-17a2e2d717c1/track.jpg","size":0},{"id":"a1786957-9780-4531-a73a-a4474ad1cc7b","type":"presentation/segment+preview","ref":"track:45210525-a5a8-4bab-a0a7-9451dfe8561e;time=T01:22:26:80F1000","mimetype":"image/jpeg","tags":{"tag":"engage-download"},"url":"https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/a1ae0961-a6b9-48ad-9f5f-eafce22f7656/track.jpg","size":0},{"id":"2a6a2641-e7fe-4c63-8b03-42458aa99da7","type":"presentation/search+preview","ref":"track:track-3","mimetype":"image/jpeg","tags":{"tag":"engage-download"},"url":"https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/attachment-3/track_1_000s_search.jpg","size":0},{"id":"3ee5c8ca-7909-42bd-b218-bd385285457f","type":"presentation/player+preview","ref":"track:track-3","mimetype":"image/jpeg","tags":{"tag":"engage-download"},"url":"https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/attachment-4/track_1_000s_player.jpg","size":0},{"id":"79f3e68a-2c1e-4520-ac9a-7141aaa1470a","type":"presentation/segment+preview","ref":"track:45210525-a5a8-4bab-a0a7-9451dfe8561e;time=T01:07:40:840F1000","mimetype":"image/jpeg","tags":{"tag":"engage-download"},"url":"https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/724f253a-afac-4c1b-b8c4-e01814af5a08/track.jpg","size":0}]},"publications":""},"acl":{"ace":[{"action":"read","allow":true,"role":"ROLE_ADMIN"},{"action":"read","allow":true,"role":"ROLE_ADMIN"},{"action":"read","allow":true,"role":"ROLE_GROUP_TUBE_COURSE_ADMINS"},{"action":"read","allow":true,"role":"ROLE_USER"},{"action":"read","allow":true,"role":"ROLE_OAUTH_USER"},{"action":"read","allow":true,"role":"ROLE_ANONYMOUS"},{"action":"write","allow":true,"role":"ROLE_ADMIN"},{"action":"write","allow":true,"role":"ROLE_ADMIN"},{"action":"write","allow":true,"role":"ROLE_GROUP_TUBE_COURSE_ADMINS"}]},"dcExtent":5414987,"dcTitle":"06 Probabilisitc Graphical Models Part 2: From Bayesian Networks to Graph Bandits","dcCreator":"Holzinger A","dcCreated":"2016-11-30T11:18:00+01:00","dcIsPartOf":"652fa592-2a86-43cb-8a9c-1c439dd00854","dcLicense":"Creative Commons 3.0: Attribution-NonCommercial-NoDerivs","ocMediapackage":"06 Probabilisitc Graphical Models Part 2: From Bayesian Networks to Graph Bandits<\/title><series>652fa592-2a86-43cb-8a9c-1c439dd00854<\/series><seriestitle>[709049] Medical Informatics<\/seriestitle><creators><creator>Holzinger A<\/creator><\/creators><license>Creative Commons 3.0: Attribution-NonCommercial-NoDerivs<\/license><media><track id=\"063565b9-8e04-432e-afdc-6f78f2b96faa\" type=\"presentation/delivery\" ref=\"track:ffd0aa9a-268d-4895-9d85-55a37afe0fa5\"><mimetype>video/mp4<\/mimetype><tags><tag>engage-download<\/tag><tag>engage-streaming<\/tag><\/tags><url>https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/7b4367f8-ded8-42dd-8367-5b5385b8021f/track.mp4<\/url><checksum type=\"md5\">d138a82adf658c4de2d7dbfcd4ce4007<\/checksum><duration>5414987<\/duration><audio id=\"audio-1\"><device/><encoder type=\"AAC (Advanced Audio Coding)\"/><channels>2<\/channels><samplingrate>44100<\/samplingrate><bitrate>64000.0<\/bitrate><\/audio><video id=\"video-1\"><device/><encoder type=\"H.264 / AVC / MPEG-4 AVC / MPEG-4 part 10\"/><bitrate>202125.0<\/bitrate><framerate>25.0<\/framerate><resolution>1280x800<\/resolution><\/video><live>false<\/live><\/track><track id=\"45210525-a5a8-4bab-a0a7-9451dfe8561e\" type=\"presentation/delivery\" ref=\"track:ffd0aa9a-268d-4895-9d85-55a37afe0fa5\" transport=\"RTMP\"><mimetype>video/mp4<\/mimetype><tags><tag>engage-download<\/tag><tag>engage-streaming<\/tag><\/tags><url>rtmp://wowza.tugraz.at/matterhorn_engage/mp4:mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/7b4367f8-ded8-42dd-8367-5b5385b8021f/track<\/url><checksum type=\"md5\">d138a82adf658c4de2d7dbfcd4ce4007<\/checksum><duration>5414987<\/duration><audio id=\"audio-1\"><device/><encoder type=\"AAC (Advanced Audio Coding)\"/><channels>2<\/channels><samplingrate>44100<\/samplingrate><bitrate>64000.0<\/bitrate><\/audio><video id=\"video-1\"><device/><encoder type=\"H.264 / AVC / MPEG-4 AVC / MPEG-4 part 10\"/><bitrate>202125.0<\/bitrate><framerate>25.0<\/framerate><resolution>1280x800<\/resolution><\/video><live>false<\/live><\/track><\/media><metadata><catalog id=\"cd56c60b-24f4-435d-adb8-92302f589a17\" type=\"dublincore/series\"><mimetype>text/xml<\/mimetype><tags><tag>archive<\/tag><tag>engage-download<\/tag><\/tags><url>https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/70a3ffa5-3225-410b-9c34-f2fa4decb6c7/70a3ffa5_3225_410b_9c34_f2fa4decb6c7.xml<\/url><checksum type=\"md5\">4caaf586bd93d89f50877dc9fa14b3dd<\/checksum><\/catalog><catalog id=\"b6c0881d-f91f-490d-b811-5014abfaf960\" type=\"dublincore/episode\"><mimetype>text/xml<\/mimetype><tags><tag>archive<\/tag><tag>engage-download<\/tag><\/tags><url>https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/4e36ced5-957c-48ea-8630-7b849984ca65/4e36ced5_957c_48ea_8630_7b849984ca65.xml<\/url><checksum type=\"md5\">5ee09835528d7136a96a56ed644c1e2a<\/checksum><\/catalog><catalog id=\"94c5ffb5-dd02-4494-b309-a71abde8ef14\" type=\"mpeg-7/text\" ref=\"track:ffd0aa9a-268d-4895-9d85-55a37afe0fa5\"><mimetype>text/xml<\/mimetype><tags><tag>engage-download<\/tag><\/tags><url>https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/9183f8ea-050f-41b3-9ba4-908e2d23a78e/slidetext.xml<\/url><\/catalog><catalog id=\"4bd8ad77-b8f5-416a-895f-18dd34257b7a\" type=\"mpeg-7/text\" ref=\"track:track-1\"><mimetype>text/xml<\/mimetype><tags><tag>archive<\/tag><tag>engage-download<\/tag><\/tags><url>https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/395f6144-4ec0-4ba4-ac5f-3df8215b07a2/catalog.xml<\/url><checksum type=\"md5\">f1597daed7f0f4d4bbd92da9c5634a16<\/checksum><\/catalog><\/metadata><attachments><attachment id=\"973e1493-1639-4a15-bbac-7b5699a858d1\" type=\"security/xacml\"><mimetype>text/xml<\/mimetype><tags><tag>archive<\/tag><\/tags><url>https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/security-policy-series/security-policy-series.xml<\/url><checksum type=\"md5\">4fe341748523f9c7035c9af3ecb78579<\/checksum><\/attachment><attachment id=\"f0b2f9e4-f4fc-4702-987d-d912e7a17f8b\" type=\"security/xacml+series\"><mimetype>text/xml<\/mimetype><tags/><url>https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/dd5b3ad7-c339-46ec-994d-40a3b9aa59d0/xacml.xml<\/url><size>0<\/size><\/attachment><attachment id=\"d6152ef5-0a06-4853-9368-85cd77d36a44\" type=\"presentation/segment+preview\" ref=\"track:45210525-a5a8-4bab-a0a7-9451dfe8561e;time=T00:21:58:560F1000\"><mimetype>image/jpeg<\/mimetype><tags><tag>engage-download<\/tag><\/tags><url>https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/1336c585-1504-49a0-a7bf-0fc9572f8c67/track.jpg<\/url><size>0<\/size><\/attachment><attachment id=\"3f16fecc-bea3-45b6-b610-213597caead5\" type=\"presentation/segment+preview\" ref=\"track:45210525-a5a8-4bab-a0a7-9451dfe8561e;time=T00:24:52:520F1000\"><mimetype>image/jpeg<\/mimetype><tags><tag>engage-download<\/tag><\/tags><url>https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/f00a235f-714a-419a-b93b-f9c733c277ee/track.jpg<\/url><size>0<\/size><\/attachment><attachment id=\"863b769a-b7a6-4904-bc51-c23ef55fdb00\" type=\"presentation/segment+preview\" ref=\"track:45210525-a5a8-4bab-a0a7-9451dfe8561e;time=T00:29:54:920F1000\"><mimetype>image/jpeg<\/mimetype><tags><tag>engage-download<\/tag><\/tags><url>https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/c579ee9b-7e66-4ac9-931d-55243d9236bc/track.jpg<\/url><size>0<\/size><\/attachment><attachment id=\"0d493e38-7153-4332-bc9d-fef87b80affa\" type=\"presentation/segment+preview\" ref=\"track:45210525-a5a8-4bab-a0a7-9451dfe8561e;time=T01:06:51:840F1000\"><mimetype>image/jpeg<\/mimetype><tags><tag>engage-download<\/tag><\/tags><url>https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/68c1787c-db47-4066-80ee-185a2205586c/track.jpg<\/url><size>0<\/size><\/attachment><attachment id=\"cb4856d6-94d6-4287-9831-c3f7de57fc42\" type=\"presentation/segment+preview\" ref=\"track:45210525-a5a8-4bab-a0a7-9451dfe8561e;time=T00:39:31:280F1000\"><mimetype>image/jpeg<\/mimetype><tags><tag>engage-download<\/tag><\/tags><url>https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/9ccda672-3ca8-40f7-a797-86c4ccc1daf6/track.jpg<\/url><size>0<\/size><\/attachment><attachment id=\"9480c549-73b0-4eb8-b108-a436230a608d\" type=\"presentation/segment+preview\" ref=\"track:45210525-a5a8-4bab-a0a7-9451dfe8561e;time=T00:49:59:120F1000\"><mimetype>image/jpeg<\/mimetype><tags><tag>engage-download<\/tag><\/tags><url>https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/c342a425-cabb-40a0-a732-535300542d22/track.jpg<\/url><size>0<\/size><\/attachment><attachment id=\"c0d5715d-1f0f-45c2-9fd6-b37faa14ed26\" type=\"presentation/segment+preview\" ref=\"track:45210525-a5a8-4bab-a0a7-9451dfe8561e;time=T00:17:14:839F1000\"><mimetype>image/jpeg<\/mimetype><tags><tag>engage-download<\/tag><\/tags><url>https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/540877d1-ab40-486f-a793-75be2b000005/track.jpg<\/url><size>0<\/size><\/attachment><attachment id=\"3ee01ed2-2025-4a76-9623-791bac9479c2\" type=\"presentation/segment+preview\" ref=\"track:45210525-a5a8-4bab-a0a7-9451dfe8561e;time=T01:28:37:360F1000\"><mimetype>image/jpeg<\/mimetype><tags><tag>engage-download<\/tag><\/tags><url>https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/71f34059-5140-44d2-bfc5-81624434cbd0/track.jpg<\/url><size>0<\/size><\/attachment><attachment id=\"fb6623e1-d80e-42c7-8cfa-43adc4e0ddc1\" type=\"presentation/segment+preview\" ref=\"track:45210525-a5a8-4bab-a0a7-9451dfe8561e;time=T00:27:43:320F1000\"><mimetype>image/jpeg<\/mimetype><tags><tag>engage-download<\/tag><\/tags><url>https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/82285407-758b-480f-ae22-6a0e64912c2a/track.jpg<\/url><size>0<\/size><\/attachment><attachment id=\"d0fc69bb-f012-4dd6-8bb0-de952e6a6290\" type=\"presentation/segment+preview\" ref=\"track:45210525-a5a8-4bab-a0a7-9451dfe8561e;time=T00:16:40:960F1000\"><mimetype>image/jpeg<\/mimetype><tags><tag>engage-download<\/tag><\/tags><url>https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/db18f079-68ae-447e-8e07-892c6773e8a2/track.jpg<\/url><size>0<\/size><\/attachment><attachment id=\"80ed5945-c89a-4382-9a3c-b30b5b41f26d\" type=\"presentation/segment+preview\" ref=\"track:45210525-a5a8-4bab-a0a7-9451dfe8561e;time=T00:14:12:360F1000\"><mimetype>image/jpeg<\/mimetype><tags><tag>engage-download<\/tag><\/tags><url>https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/4be1e541-ff90-44b7-8f84-c116cccc87b1/track.jpg<\/url><size>0<\/size><\/attachment><attachment id=\"e80c368c-2240-4ec7-8ac8-278724c59962\" type=\"presentation/segment+preview\" ref=\"track:45210525-a5a8-4bab-a0a7-9451dfe8561e;time=T01:05:08:880F1000\"><mimetype>image/jpeg<\/mimetype><tags><tag>engage-download<\/tag><\/tags><url>https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/835ca2af-7235-4911-aa10-c3442fcbf6a6/track.jpg<\/url><size>0<\/size><\/attachment><attachment id=\"25024805-77f0-437a-9045-d4fdc08181f6\" type=\"presentation/segment+preview\" ref=\"track:45210525-a5a8-4bab-a0a7-9451dfe8561e;time=T00:12:20:280F1000\"><mimetype>image/jpeg<\/mimetype><tags><tag>engage-download<\/tag><\/tags><url>https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/15d297a9-96b7-41dd-894d-6e4fa92d1a84/track.jpg<\/url><size>0<\/size><\/attachment><attachment id=\"ff43a14f-c068-41ad-ba53-5b94a4758a58\" type=\"presentation/segment+preview\" ref=\"track:45210525-a5a8-4bab-a0a7-9451dfe8561e;time=T01:07:23:680F1000\"><mimetype>image/jpeg<\/mimetype><tags><tag>engage-download<\/tag><\/tags><url>https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/1e052c0a-5870-4dbe-ad1b-3c15715ccbd2/track.jpg<\/url><size>0<\/size><\/attachment><attachment id=\"b50bdcd3-9fef-425f-ba28-d03488e5c01c\" type=\"presentation/segment+preview\" ref=\"track:45210525-a5a8-4bab-a0a7-9451dfe8561e;time=T00:43:10:920F1000\"><mimetype>image/jpeg<\/mimetype><tags><tag>engage-download<\/tag><\/tags><url>https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/e9325995-b460-4c65-9c8a-72f0e3972b25/track.jpg<\/url><size>0<\/size><\/attachment><attachment id=\"6272e0b8-5f77-43c2-a2ae-85c64813deff\" type=\"presentation/segment+preview\" ref=\"track:45210525-a5a8-4bab-a0a7-9451dfe8561e;time=T01:26:33:920F1000\"><mimetype>image/jpeg<\/mimetype><tags><tag>engage-download<\/tag><\/tags><url>https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/e3960148-e0b0-45c3-84ac-b8fc8bcd4ea1/track.jpg<\/url><size>0<\/size><\/attachment><attachment id=\"b0883db4-e094-4689-b42c-9a3e6ddd4d43\" type=\"presentation/segment+preview\" ref=\"track:45210525-a5a8-4bab-a0a7-9451dfe8561e;time=T00:20:56:40F1000\"><mimetype>image/jpeg<\/mimetype><tags><tag>engage-download<\/tag><\/tags><url>https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/a4fd407c-ab7b-4a5e-8434-7dcb1100fd5e/track.jpg<\/url><size>0<\/size><\/attachment><attachment id=\"f6ddd8d0-9557-421f-8fa2-0a11d3d51eab\" type=\"presentation/segment+preview\" ref=\"track:45210525-a5a8-4bab-a0a7-9451dfe8561e;time=T01:24:53:80F1000\"><mimetype>image/jpeg<\/mimetype><tags><tag>engage-download<\/tag><\/tags><url>https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/e3f0f740-6fc8-415b-bba3-6a248bc21de2/track.jpg<\/url><size>0<\/size><\/attachment><attachment id=\"36d183af-6fe9-4a69-9d51-e43dd53325f2\" type=\"presentation/segment+preview\" ref=\"track:45210525-a5a8-4bab-a0a7-9451dfe8561e;time=T00:52:04:880F1000\"><mimetype>image/jpeg<\/mimetype><tags><tag>engage-download<\/tag><\/tags><url>https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/d6759621-7816-423d-9969-dd4cccb1b4ba/track.jpg<\/url><size>0<\/size><\/attachment><attachment id=\"4ed36e18-5feb-495c-8c33-7e2ebe094952\" type=\"presentation/segment+preview\" ref=\"track:45210525-a5a8-4bab-a0a7-9451dfe8561e;time=T00:12:13:800F1000\"><mimetype>image/jpeg<\/mimetype><tags><tag>engage-download<\/tag><\/tags><url>https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/15e202d1-27b0-496e-ac89-ca801fca8e2f/track.jpg<\/url><size>0<\/size><\/attachment><attachment id=\"fee30829-93c6-4174-ab7e-aeb51218eec1\" type=\"presentation/segment+preview\" ref=\"track:45210525-a5a8-4bab-a0a7-9451dfe8561e;time=T00:24:37:760F1000\"><mimetype>image/jpeg<\/mimetype><tags><tag>engage-download<\/tag><\/tags><url>https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/6caf36e0-10ef-496a-adc7-2c4bc89ef2bd/track.jpg<\/url><size>0<\/size><\/attachment><attachment id=\"3ce34bd0-a5c1-4765-9141-083f32b5ba4d\" type=\"presentation/segment+preview\" ref=\"track:45210525-a5a8-4bab-a0a7-9451dfe8561e;time=T01:02:37:960F1000\"><mimetype>image/jpeg<\/mimetype><tags><tag>engage-download<\/tag><\/tags><url>https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/17e75605-f22c-49d6-9fbc-c0f4900bcfbc/track.jpg<\/url><size>0<\/size><\/attachment><attachment id=\"c7da78b9-63ef-47b4-ba14-05ae686ad8df\" type=\"presentation/segment+preview\" ref=\"track:45210525-a5a8-4bab-a0a7-9451dfe8561e;time=T01:04:28:600F1000\"><mimetype>image/jpeg<\/mimetype><tags><tag>engage-download<\/tag><\/tags><url>https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/f0de732e-5ddd-4caf-ab64-ba2e7939cb70/track.jpg<\/url><size>0<\/size><\/attachment><attachment id=\"a78a5636-c0d8-48fc-8890-d30851e28933\" type=\"presentation/segment+preview\" ref=\"track:45210525-a5a8-4bab-a0a7-9451dfe8561e;time=T01:25:52:560F1000\"><mimetype>image/jpeg<\/mimetype><tags><tag>engage-download<\/tag><\/tags><url>https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/bf4d1411-16c2-48b3-b9eb-4e720366858f/track.jpg<\/url><size>0<\/size><\/attachment><attachment id=\"7de914dd-543d-4840-9295-054a79829d9a\" type=\"presentation/segment+preview\" ref=\"track:45210525-a5a8-4bab-a0a7-9451dfe8561e;time=T00:28:40:360F1000\"><mimetype>image/jpeg<\/mimetype><tags><tag>engage-download<\/tag><\/tags><url>https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/6446e2b3-f7a8-4e64-9df1-46d63281b45b/track.jpg<\/url><size>0<\/size><\/attachment><attachment id=\"50f9aad9-c53e-4095-844a-f82781db8d58\" type=\"presentation/segment+preview\" ref=\"track:45210525-a5a8-4bab-a0a7-9451dfe8561e;time=T01:14:00:600F1000\"><mimetype>image/jpeg<\/mimetype><tags><tag>engage-download<\/tag><\/tags><url>https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/e2acd983-4a95-424d-8ad9-73111a1d8178/track.jpg<\/url><size>0<\/size><\/attachment><attachment id=\"7defadb9-2ca1-48b0-883e-57995fa247fa\" type=\"presentation/segment+preview\" ref=\"track:45210525-a5a8-4bab-a0a7-9451dfe8561e;time=T00:32:12:280F1000\"><mimetype>image/jpeg<\/mimetype><tags><tag>engage-download<\/tag><\/tags><url>https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/b29a2412-d7de-4f97-8434-0025dcb4a7a3/track.jpg<\/url><size>0<\/size><\/attachment><attachment id=\"ad14abad-e5cc-4d90-8fdc-fc46968ebaf7\" type=\"presentation/segment+preview\" ref=\"track:45210525-a5a8-4bab-a0a7-9451dfe8561e;time=T01:20:21:40F1000\"><mimetype>image/jpeg<\/mimetype><tags><tag>engage-download<\/tag><\/tags><url>https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/98f14e9e-0bd5-4814-accc-27dd0c253161/track.jpg<\/url><size>0<\/size><\/attachment><attachment id=\"097e20ed-aa55-4a8b-a26e-c3c7db3f1921\" type=\"presentation/segment+preview\" ref=\"track:45210525-a5a8-4bab-a0a7-9451dfe8561e;time=T00:45:48:400F1000\"><mimetype>image/jpeg<\/mimetype><tags><tag>engage-download<\/tag><\/tags><url>https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/6ec9c41f-f1dd-41d8-966c-cc852d005075/track.jpg<\/url><size>0<\/size><\/attachment><attachment id=\"3c86b5ca-4260-4b52-a07a-864bd48965b0\" type=\"presentation/segment+preview\" ref=\"track:45210525-a5a8-4bab-a0a7-9451dfe8561e;time=T00:01:54:560F1000\"><mimetype>image/jpeg<\/mimetype><tags><tag>engage-download<\/tag><\/tags><url>https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/274b400c-9e92-4b98-8b8a-52c3b852ad03/track.jpg<\/url><size>0<\/size><\/attachment><attachment id=\"3b56ec9d-62fa-46ec-abe5-70becc1bbe92\" type=\"presentation/segment+preview\" ref=\"track:45210525-a5a8-4bab-a0a7-9451dfe8561e;time=T00:08:17:360F1000\"><mimetype>image/jpeg<\/mimetype><tags><tag>engage-download<\/tag><\/tags><url>https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/6e951bc3-5e85-4aa8-a0b2-5238edd24b13/track.jpg<\/url><size>0<\/size><\/attachment><attachment id=\"5a477fec-238c-4117-82f0-18ce3dba34d4\" type=\"presentation/segment+preview\" ref=\"track:45210525-a5a8-4bab-a0a7-9451dfe8561e;time=T01:03:05:600F1000\"><mimetype>image/jpeg<\/mimetype><tags><tag>engage-download<\/tag><\/tags><url>https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/82b5188e-6bfa-463f-a815-e6bfc091c6eb/track.jpg<\/url><size>0<\/size><\/attachment><attachment id=\"f8ec2f9f-3bba-4f61-b56f-fb740a7427cb\" type=\"presentation/segment+preview\" ref=\"track:45210525-a5a8-4bab-a0a7-9451dfe8561e;time=T00:09:43:480F1000\"><mimetype>image/jpeg<\/mimetype><tags><tag>engage-download<\/tag><\/tags><url>https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/e8c5b1bf-d1de-4806-b6ac-a26482b5deac/track.jpg<\/url><size>0<\/size><\/attachment><attachment id=\"db089d92-a64b-4dd0-9b97-42bc5d8bc434\" type=\"presentation/segment+preview\" ref=\"track:45210525-a5a8-4bab-a0a7-9451dfe8561e;time=T00:33:59:720F1000\"><mimetype>image/jpeg<\/mimetype><tags><tag>engage-download<\/tag><\/tags><url>https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/39335258-3766-42e2-b74b-f21bfc4b5a12/track.jpg<\/url><size>0<\/size><\/attachment><attachment id=\"8689b77a-2344-461e-8032-d6d3cc58a2df\" type=\"presentation/segment+preview\" ref=\"track:45210525-a5a8-4bab-a0a7-9451dfe8561e;time=T00:34:36:40F1000\"><mimetype>image/jpeg<\/mimetype><tags><tag>engage-download<\/tag><\/tags><url>https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/be112ef5-6009-4bc3-a476-4ecb3a9d77a0/track.jpg<\/url><size>0<\/size><\/attachment><attachment id=\"4e9e957f-0d56-4861-bde9-2dc207ed2fdf\" type=\"presentation/segment+preview\" ref=\"track:45210525-a5a8-4bab-a0a7-9451dfe8561e;time=T00:19:19:480F1000\"><mimetype>image/jpeg<\/mimetype><tags><tag>engage-download<\/tag><\/tags><url>https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/a80a39c5-c445-4304-882f-6887423f6e96/track.jpg<\/url><size>0<\/size><\/attachment><attachment id=\"4f7d419a-1af0-4d24-9af2-b6e528c8facc\" type=\"presentation/segment+preview\" ref=\"track:45210525-a5a8-4bab-a0a7-9451dfe8561e;time=T00:09:16:200F1000\"><mimetype>image/jpeg<\/mimetype><tags><tag>engage-download<\/tag><\/tags><url>https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/3415228d-5667-4ed7-8379-69d95befa033/track.jpg<\/url><size>0<\/size><\/attachment><attachment id=\"15fc84d6-d6ad-4e0f-aa06-33c79a9ce129\" type=\"presentation/segment+preview\" ref=\"track:45210525-a5a8-4bab-a0a7-9451dfe8561e;time=T01:04:55:280F1000\"><mimetype>image/jpeg<\/mimetype><tags><tag>engage-download<\/tag><\/tags><url>https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/ca1ad350-3192-41f2-b2be-734b1bba7373/track.jpg<\/url><size>0<\/size><\/attachment><attachment id=\"b1ba5ad4-88fb-4ec1-82eb-3f74e5d9b39f\" type=\"presentation/segment+preview\" ref=\"track:45210525-a5a8-4bab-a0a7-9451dfe8561e;time=T01:24:59:640F1000\"><mimetype>image/jpeg<\/mimetype><tags><tag>engage-download<\/tag><\/tags><url>https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/7020a465-0d40-4eb6-8454-6ecbcd819a40/track.jpg<\/url><size>0<\/size><\/attachment><attachment id=\"bb5a0700-1b5b-4bdb-b06a-842ea9355d10\" type=\"presentation/segment+preview\" ref=\"track:45210525-a5a8-4bab-a0a7-9451dfe8561e;time=T00:49:03:680F1000\"><mimetype>image/jpeg<\/mimetype><tags><tag>engage-download<\/tag><\/tags><url>https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/6dfc7169-200b-4084-b780-bfd15c6e3651/track.jpg<\/url><size>0<\/size><\/attachment><attachment id=\"3f235ee8-fd35-4678-ad0c-90faaeb3b88a\" type=\"presentation/segment+preview\" ref=\"track:45210525-a5a8-4bab-a0a7-9451dfe8561e;time=T01:28:07:200F1000\"><mimetype>image/jpeg<\/mimetype><tags><tag>engage-download<\/tag><\/tags><url>https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/a1838e06-c47e-42bc-a268-38734a107ef3/track.jpg<\/url><size>0<\/size><\/attachment><attachment id=\"cc4ede90-3c22-4a0b-ba64-129cb94c97e7\" type=\"presentation/segment+preview\" ref=\"track:45210525-a5a8-4bab-a0a7-9451dfe8561e;time=T00:56:34:40F1000\"><mimetype>image/jpeg<\/mimetype><tags><tag>engage-download<\/tag><\/tags><url>https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/eee918ba-be51-4bd6-97a0-ab7f740bd4c7/track.jpg<\/url><size>0<\/size><\/attachment><attachment id=\"b045c314-1484-457d-bdf9-db372e84e80a\" type=\"presentation/segment+preview\" ref=\"track:45210525-a5a8-4bab-a0a7-9451dfe8561e;time=T00:00:00:0F1000\"><mimetype>image/jpeg<\/mimetype><tags><tag>engage-download<\/tag><\/tags><url>https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/d2128d70-db12-49ec-99e4-3f00f21cc51f/track.jpg<\/url><size>0<\/size><\/attachment><attachment id=\"f54e9273-e13d-45e7-98e3-babbc856ebc5\" type=\"presentation/segment+preview\" ref=\"track:45210525-a5a8-4bab-a0a7-9451dfe8561e;time=T00:48:07:520F1000\"><mimetype>image/jpeg<\/mimetype><tags><tag>engage-download<\/tag><\/tags><url>https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/dffc5f4a-3b7d-4af9-9753-04151aaca684/track.jpg<\/url><size>0<\/size><\/attachment><attachment id=\"efcc30fd-9f83-43cd-8ffe-d72462ce8faa\" type=\"presentation/segment+preview\" ref=\"track:45210525-a5a8-4bab-a0a7-9451dfe8561e;time=T00:09:32:920F1000\"><mimetype>image/jpeg<\/mimetype><tags><tag>engage-download<\/tag><\/tags><url>https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/5891df7b-cac0-4cf9-b64f-d1049234811a/track.jpg<\/url><size>0<\/size><\/attachment><attachment id=\"f3c428be-b543-4e25-a4b5-7edd15831998\" type=\"presentation/segment+preview\" ref=\"track:45210525-a5a8-4bab-a0a7-9451dfe8561e;time=T00:53:32:200F1000\"><mimetype>image/jpeg<\/mimetype><tags><tag>engage-download<\/tag><\/tags><url>https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/8c57954f-4451-45f0-98ba-131bb2f55272/track.jpg<\/url><size>0<\/size><\/attachment><attachment id=\"8aba5bc7-3ed6-4e84-aee1-7608e1dde532\" type=\"presentation/segment+preview\" ref=\"track:45210525-a5a8-4bab-a0a7-9451dfe8561e;time=T01:24:28:880F1000\"><mimetype>image/jpeg<\/mimetype><tags><tag>engage-download<\/tag><\/tags><url>https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/628b2a11-514b-4fa3-8ba9-9aca78ad4e2e/track.jpg<\/url><size>0<\/size><\/attachment><attachment id=\"7538d46d-7538-4ecf-ac09-1ab6ecb1d9e6\" type=\"presenter/player+preview\"><mimetype>image/png<\/mimetype><tags><tag>archive<\/tag><tag>engage-download<\/tag><\/tags><url>https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/e7e94061-3cfb-452e-a1c4-e6a24dd52fc5/coverimage.png<\/url><size>0<\/size><\/attachment><attachment id=\"bce001ac-79e8-449d-aa6a-b1e1785df3e1\" type=\"presentation/segment+preview\" ref=\"track:45210525-a5a8-4bab-a0a7-9451dfe8561e;time=T01:05:48:880F1000\"><mimetype>image/jpeg<\/mimetype><tags><tag>engage-download<\/tag><\/tags><url>https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/9c3c7f8a-c597-4d07-bc3b-62bae8b79223/track.jpg<\/url><size>0<\/size><\/attachment><attachment id=\"44aad8d9-6ee0-4422-a6c3-070ea404ce0d\" type=\"presentation/segment+preview\" ref=\"track:45210525-a5a8-4bab-a0a7-9451dfe8561e;time=T01:18:14:240F1000\"><mimetype>image/jpeg<\/mimetype><tags><tag>engage-download<\/tag><\/tags><url>https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/3016dfa2-6f63-4838-905a-56503c1ddea3/track.jpg<\/url><size>0<\/size><\/attachment><attachment id=\"818531d9-0657-401d-a8e0-921fa01d5639\" type=\"presentation/segment+preview\" ref=\"track:45210525-a5a8-4bab-a0a7-9451dfe8561e;time=T01:21:19:240F1000\"><mimetype>image/jpeg<\/mimetype><tags><tag>engage-download<\/tag><\/tags><url>https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/2d278a9a-1a23-449f-8cdd-718ef36b6bf5/track.jpg<\/url><size>0<\/size><\/attachment><attachment id=\"f7652451-f6fd-4a68-8d00-dada76109346\" type=\"presentation/segment+preview\" ref=\"track:45210525-a5a8-4bab-a0a7-9451dfe8561e;time=T01:12:05:0F1000\"><mimetype>image/jpeg<\/mimetype><tags><tag>engage-download<\/tag><\/tags><url>https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/d7ec3d3e-0242-4bb6-ae8e-10e8cc9d8b47/track.jpg<\/url><size>0<\/size><\/attachment><attachment id=\"6983e8ae-51b4-41e0-981a-9e6ea7ab5179\" type=\"presentation/segment+preview\" ref=\"track:45210525-a5a8-4bab-a0a7-9451dfe8561e;time=T01:01:26:560F1000\"><mimetype>image/jpeg<\/mimetype><tags><tag>engage-download<\/tag><\/tags><url>https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/661b3521-6399-4131-b8d2-bfd5c1d47441/track.jpg<\/url><size>0<\/size><\/attachment><attachment id=\"6027a2c7-db2e-4005-9645-a8a5e65cd9f6\" type=\"presentation/segment+preview\" ref=\"track:45210525-a5a8-4bab-a0a7-9451dfe8561e;time=T01:13:20:400F1000\"><mimetype>image/jpeg<\/mimetype><tags><tag>engage-download<\/tag><\/tags><url>https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/63a39edf-fc5e-43b7-8b22-dad80bdda431/track.jpg<\/url><size>0<\/size><\/attachment><attachment id=\"12f87a68-a57b-4949-a25a-4640fc043669\" type=\"presentation/segment+preview\" ref=\"track:45210525-a5a8-4bab-a0a7-9451dfe8561e;time=T00:20:14:240F1000\"><mimetype>image/jpeg<\/mimetype><tags><tag>engage-download<\/tag><\/tags><url>https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/f22280d2-d58d-4f7b-a45d-b93526d70fbe/track.jpg<\/url><size>0<\/size><\/attachment><attachment id=\"5add9872-dd04-433b-ba69-84506a3bc438\" type=\"presentation/segment+preview\" ref=\"track:45210525-a5a8-4bab-a0a7-9451dfe8561e;time=T00:54:42:680F1000\"><mimetype>image/jpeg<\/mimetype><tags><tag>engage-download<\/tag><\/tags><url>https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/5f67c6bd-1941-4052-a696-96aa17daefbc/track.jpg<\/url><size>0<\/size><\/attachment><attachment id=\"5ef67b12-360e-41d3-955e-2531c918e187\" type=\"presentation/segment+preview\" ref=\"track:45210525-a5a8-4bab-a0a7-9451dfe8561e;time=T00:54:51:960F1000\"><mimetype>image/jpeg<\/mimetype><tags><tag>engage-download<\/tag><\/tags><url>https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/35a92b3f-2e1e-4308-9c90-0a2b39f6bb51/track.jpg<\/url><size>0<\/size><\/attachment><attachment id=\"72b38d5e-8f37-4319-91fc-5807e3662ab4\" type=\"presentation/segment+preview\" ref=\"track:45210525-a5a8-4bab-a0a7-9451dfe8561e;time=T00:38:53:920F1000\"><mimetype>image/jpeg<\/mimetype><tags><tag>engage-download<\/tag><\/tags><url>https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/9af7c54f-3da7-49c3-852e-bf3b66f6aded/track.jpg<\/url><size>0<\/size><\/attachment><attachment id=\"97a40235-0821-4232-8643-655c868636d0\" type=\"presentation/segment+preview\" ref=\"track:45210525-a5a8-4bab-a0a7-9451dfe8561e;time=T00:41:14:320F1000\"><mimetype>image/jpeg<\/mimetype><tags><tag>engage-download<\/tag><\/tags><url>https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/a0c59d71-959f-4c49-ab51-5e462bf4c221/track.jpg<\/url><size>0<\/size><\/attachment><attachment id=\"92744892-71fc-4553-acc5-944b2411df4e\" type=\"presentation/segment+preview\" ref=\"track:45210525-a5a8-4bab-a0a7-9451dfe8561e;time=T01:06:09:360F1000\"><mimetype>image/jpeg<\/mimetype><tags><tag>engage-download<\/tag><\/tags><url>https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/33c4b33a-8a0a-45f8-9086-354b82a36b26/track.jpg<\/url><size>0<\/size><\/attachment><attachment id=\"04da209b-bf32-4bb9-b1e6-410eed901c4f\" type=\"presentation/segment+preview\" ref=\"track:45210525-a5a8-4bab-a0a7-9451dfe8561e;time=T01:27:30:560F1000\"><mimetype>image/jpeg<\/mimetype><tags><tag>engage-download<\/tag><\/tags><url>https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/7a161758-64bf-4e3e-97f9-825009822213/track.jpg<\/url><size>0<\/size><\/attachment><attachment id=\"e2de28e8-9bac-4c38-9ede-e20c4843dbf2\" type=\"presentation/segment+preview\" ref=\"track:45210525-a5a8-4bab-a0a7-9451dfe8561e;time=T00:42:31:360F1000\"><mimetype>image/jpeg<\/mimetype><tags><tag>engage-download<\/tag><\/tags><url>https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/f0952a7a-d650-4656-90f4-bb7e7dc5e949/track.jpg<\/url><size>0<\/size><\/attachment><attachment id=\"d05d8cf6-e4d7-4374-b1b3-2651cde1d98e\" type=\"presentation/segment+preview\" ref=\"track:45210525-a5a8-4bab-a0a7-9451dfe8561e;time=T01:09:20:799F1000\"><mimetype>image/jpeg<\/mimetype><tags><tag>engage-download<\/tag><\/tags><url>https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/4cef5745-e19a-4571-9b11-84b3948d8d8e/track.jpg<\/url><size>0<\/size><\/attachment><attachment id=\"b84c1dab-06fc-4f11-8c85-3335fc042429\" type=\"presentation/segment+preview\" ref=\"track:45210525-a5a8-4bab-a0a7-9451dfe8561e;time=T00:33:10:880F1000\"><mimetype>image/jpeg<\/mimetype><tags><tag>engage-download<\/tag><\/tags><url>https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/5b258709-ba4f-46e0-aae6-49a61a64e25b/track.jpg<\/url><size>0<\/size><\/attachment><attachment id=\"d841ccdc-4336-405a-ac19-0a381c327709\" type=\"presentation/segment+preview\" ref=\"track:45210525-a5a8-4bab-a0a7-9451dfe8561e;time=T00:26:44:960F1000\"><mimetype>image/jpeg<\/mimetype><tags><tag>engage-download<\/tag><\/tags><url>https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/519507a6-eec8-4782-b57d-9d09f351975e/track.jpg<\/url><size>0<\/size><\/attachment><attachment id=\"19acf831-1ad7-40d4-ab04-b5caa5931973\" type=\"presentation/segment+preview\" ref=\"track:45210525-a5a8-4bab-a0a7-9451dfe8561e;time=T00:31:00:480F1000\"><mimetype>image/jpeg<\/mimetype><tags><tag>engage-download<\/tag><\/tags><url>https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/83fb9e15-4f84-48c0-8ef9-0ac99e2300cc/track.jpg<\/url><size>0<\/size><\/attachment><attachment id=\"ff5278dd-824c-4e92-beaf-7e57ffcc5c90\" type=\"presentation/segment+preview\" ref=\"track:45210525-a5a8-4bab-a0a7-9451dfe8561e;time=T01:25:09:360F1000\"><mimetype>image/jpeg<\/mimetype><tags><tag>engage-download<\/tag><\/tags><url>https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/2f98a0d2-63ce-4bd0-bed8-56286b2bc93c/track.jpg<\/url><size>0<\/size><\/attachment><attachment id=\"f608aa9b-73ca-4986-8d14-b970b4696f44\" type=\"presentation/segment+preview\" ref=\"track:45210525-a5a8-4bab-a0a7-9451dfe8561e;time=T00:51:45:480F1000\"><mimetype>image/jpeg<\/mimetype><tags><tag>engage-download<\/tag><\/tags><url>https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/53f08c19-ebc5-4ce7-9ddb-2681eb42c00c/track.jpg<\/url><size>0<\/size><\/attachment><attachment id=\"d3745f34-1d3f-4738-914a-68ef9f1b6901\" type=\"presentation/segment+preview\" ref=\"track:45210525-a5a8-4bab-a0a7-9451dfe8561e;time=T01:04:37:400F1000\"><mimetype>image/jpeg<\/mimetype><tags><tag>engage-download<\/tag><\/tags><url>https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/addb0b35-fee3-4ef9-9680-e9d913c71982/track.jpg<\/url><size>0<\/size><\/attachment><attachment id=\"3480a20a-c8c4-4eee-9a63-73a75a2faa73\" type=\"presentation/segment+preview\" ref=\"track:45210525-a5a8-4bab-a0a7-9451dfe8561e;time=T01:27:38:960F1000\"><mimetype>image/jpeg<\/mimetype><tags><tag>engage-download<\/tag><\/tags><url>https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/8451a69f-11ad-4902-a25a-d412b52cf810/track.jpg<\/url><size>0<\/size><\/attachment><attachment id=\"d4b2a422-ffca-4690-80d6-84dbd79e33fe\" type=\"presentation/segment+preview\" ref=\"track:45210525-a5a8-4bab-a0a7-9451dfe8561e;time=T01:29:58:640F1000\"><mimetype>image/jpeg<\/mimetype><tags><tag>engage-download<\/tag><\/tags><url>https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/8e98bc46-c08d-4e0a-a37b-f8b7e96199f1/track.jpg<\/url><size>0<\/size><\/attachment><attachment id=\"346c06cc-569b-47e3-b84d-b8b93aa8ab38\" type=\"presentation/segment+preview\" ref=\"track:45210525-a5a8-4bab-a0a7-9451dfe8561e;time=T00:46:28:320F1000\"><mimetype>image/jpeg<\/mimetype><tags><tag>engage-download<\/tag><\/tags><url>https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/ffac8015-7952-4061-a9c5-00c95b5de942/track.jpg<\/url><size>0<\/size><\/attachment><attachment id=\"5abe3c0c-6a1b-4ab0-8818-f6866bd82ab8\" type=\"presentation/segment+preview\" ref=\"track:45210525-a5a8-4bab-a0a7-9451dfe8561e;time=T00:43:16:200F1000\"><mimetype>image/jpeg<\/mimetype><tags><tag>engage-download<\/tag><\/tags><url>https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/e62c918e-485c-45c6-9e7f-742c7e1446c4/track.jpg<\/url><size>0<\/size><\/attachment><attachment id=\"13bdb656-6534-41bd-8528-4aefce7646aa\" type=\"presentation/segment+preview\" ref=\"track:45210525-a5a8-4bab-a0a7-9451dfe8561e;time=T01:08:07:920F1000\"><mimetype>image/jpeg<\/mimetype><tags><tag>engage-download<\/tag><\/tags><url>https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/f16d5787-f7b0-4635-8675-8ea5c56411c1/track.jpg<\/url><size>0<\/size><\/attachment><attachment id=\"714ded71-ac02-4b26-9369-05670137a278\" type=\"presentation/segment+preview\" ref=\"track:45210525-a5a8-4bab-a0a7-9451dfe8561e;time=T00:26:12:0F1000\"><mimetype>image/jpeg<\/mimetype><tags><tag>engage-download<\/tag><\/tags><url>https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/e376250c-5db2-4426-b393-bbc4e7a2d19a/track.jpg<\/url><size>0<\/size><\/attachment><attachment id=\"240efa5e-dbb8-4fcf-a933-8306e345a8a2\" type=\"presentation/segment+preview\" ref=\"track:45210525-a5a8-4bab-a0a7-9451dfe8561e;time=T00:05:53:160F1000\"><mimetype>image/jpeg<\/mimetype><tags><tag>engage-download<\/tag><\/tags><url>https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/d5e51fdc-4d7c-42ce-a3de-dae9ffdb8103/track.jpg<\/url><size>0<\/size><\/attachment><attachment id=\"db521ab6-3d3c-4980-8030-e155b7afb3a7\" type=\"presentation/segment+preview\" ref=\"track:45210525-a5a8-4bab-a0a7-9451dfe8561e;time=T00:23:42:160F1000\"><mimetype>image/jpeg<\/mimetype><tags><tag>engage-download<\/tag><\/tags><url>https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/a2295e0a-f7ef-41fd-847b-40679ae28d5d/track.jpg<\/url><size>0<\/size><\/attachment><attachment id=\"d1b9dbb3-b8bf-4772-a9c2-44367f60bb15\" type=\"presentation/segment+preview\" ref=\"track:45210525-a5a8-4bab-a0a7-9451dfe8561e;time=T00:35:52:0F1000\"><mimetype>image/jpeg<\/mimetype><tags><tag>engage-download<\/tag><\/tags><url>https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/f9ba87ff-9ee0-4adc-a383-1b00646b7527/track.jpg<\/url><size>0<\/size><\/attachment><attachment id=\"634d7de0-430a-497c-99ee-239a4aa3ff53\" type=\"presentation/segment+preview\" ref=\"track:45210525-a5a8-4bab-a0a7-9451dfe8561e;time=T00:36:44:280F1000\"><mimetype>image/jpeg<\/mimetype><tags><tag>engage-download<\/tag><\/tags><url>https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/91059181-d401-4617-b548-81e0723871da/track.jpg<\/url><size>0<\/size><\/attachment><attachment id=\"ba5ef2e4-86c6-4d27-9c76-627b329265f5\" type=\"presentation/segment+preview\" ref=\"track:45210525-a5a8-4bab-a0a7-9451dfe8561e;time=T01:08:43:120F1000\"><mimetype>image/jpeg<\/mimetype><tags><tag>engage-download<\/tag><\/tags><url>https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/13803432-0117-4425-9cca-21ef610c3271/track.jpg<\/url><size>0<\/size><\/attachment><attachment id=\"8df8e840-81ff-49a5-abb2-e2bd5a3f28c0\" type=\"presentation/segment+preview\" ref=\"track:45210525-a5a8-4bab-a0a7-9451dfe8561e;time=T01:07:15:320F1000\"><mimetype>image/jpeg<\/mimetype><tags><tag>engage-download<\/tag><\/tags><url>https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/9abd0d03-0891-458b-8006-0a911695502d/track.jpg<\/url><size>0<\/size><\/attachment><attachment id=\"7e35f66c-e6ff-4936-8be3-92c4b1e9de42\" type=\"presentation/segment+preview\" ref=\"track:45210525-a5a8-4bab-a0a7-9451dfe8561e;time=T00:35:08:320F1000\"><mimetype>image/jpeg<\/mimetype><tags><tag>engage-download<\/tag><\/tags><url>https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/f1acca10-7339-48f3-b8a8-45d235c6b0c9/track.jpg<\/url><size>0<\/size><\/attachment><attachment id=\"69ed39a7-86d8-40cf-9ae8-03292ae5a785\" type=\"presentation/segment+preview\" ref=\"track:45210525-a5a8-4bab-a0a7-9451dfe8561e;time=T00:57:31:920F1000\"><mimetype>image/jpeg<\/mimetype><tags><tag>engage-download<\/tag><\/tags><url>https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/57a64a0d-b0dc-4640-8237-a4df5ef3f941/track.jpg<\/url><size>0<\/size><\/attachment><attachment id=\"f0723b44-bc94-4997-b788-9150169dca73\" type=\"presentation/segment+preview\" ref=\"track:45210525-a5a8-4bab-a0a7-9451dfe8561e;time=T00:59:43:880F1000\"><mimetype>image/jpeg<\/mimetype><tags><tag>engage-download<\/tag><\/tags><url>https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/de1e857b-3297-4b8e-8197-4734ae81cf41/track.jpg<\/url><size>0<\/size><\/attachment><attachment id=\"55c49d99-e3df-40ca-a131-fb5df173440d\" type=\"presentation/segment+preview\" ref=\"track:45210525-a5a8-4bab-a0a7-9451dfe8561e;time=T01:28:31:560F1000\"><mimetype>image/jpeg<\/mimetype><tags><tag>engage-download<\/tag><\/tags><url>https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/a521675a-8f35-40a8-acd3-8fe6991f0a3f/track.jpg<\/url><size>0<\/size><\/attachment><attachment id=\"469ec150-b26c-4ca1-a151-af9d7cea92e1\" type=\"presentation/segment+preview\" ref=\"track:45210525-a5a8-4bab-a0a7-9451dfe8561e;time=T00:35:00:960F1000\"><mimetype>image/jpeg<\/mimetype><tags><tag>engage-download<\/tag><\/tags><url>https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/ea79344c-865a-4898-970a-5f1caead5f7a/track.jpg<\/url><size>0<\/size><\/attachment><attachment id=\"52cdd696-38e5-44c4-9d7a-e4f3a69738a3\" type=\"presentation/segment+preview\" ref=\"track:45210525-a5a8-4bab-a0a7-9451dfe8561e;time=T00:25:20:520F1000\"><mimetype>image/jpeg<\/mimetype><tags><tag>engage-download<\/tag><\/tags><url>https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/6d5a6667-c1ad-476b-9d18-56ec9de25081/track.jpg<\/url><size>0<\/size><\/attachment><attachment id=\"87dbf3ff-624a-443e-8af5-417551f26a51\" type=\"presentation/segment+preview\" ref=\"track:45210525-a5a8-4bab-a0a7-9451dfe8561e;time=T00:19:25:880F1000\"><mimetype>image/jpeg<\/mimetype><tags><tag>engage-download<\/tag><\/tags><url>https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/ef3d6ebe-737c-42a6-aedd-9fc8ee781d11/track.jpg<\/url><size>0<\/size><\/attachment><attachment id=\"a08de618-f4c2-4db4-8fb8-33456baf6298\" type=\"presentation/segment+preview\" ref=\"track:45210525-a5a8-4bab-a0a7-9451dfe8561e;time=T01:24:05:480F1000\"><mimetype>image/jpeg<\/mimetype><tags><tag>engage-download<\/tag><\/tags><url>https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/1b3fc3a1-9d79-42f4-9d9c-9809b2121efa/track.jpg<\/url><size>0<\/size><\/attachment><attachment id=\"dfa2848a-b93e-42ae-aac5-b3b39d65f4b6\" type=\"presentation/segment+preview\" ref=\"track:45210525-a5a8-4bab-a0a7-9451dfe8561e;time=T00:50:17:480F1000\"><mimetype>image/jpeg<\/mimetype><tags><tag>engage-download<\/tag><\/tags><url>https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/9724eb4c-9371-41a2-84b0-b216a65730fa/track.jpg<\/url><size>0<\/size><\/attachment><attachment id=\"f7bd3d95-f4c4-4779-9789-e7fc5f61cbdd\" type=\"presentation/segment+preview\" ref=\"track:45210525-a5a8-4bab-a0a7-9451dfe8561e;time=T01:12:50:680F1000\"><mimetype>image/jpeg<\/mimetype><tags><tag>engage-download<\/tag><\/tags><url>https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/cd276b03-f8c6-443b-b416-4d4ed7385e1b/track.jpg<\/url><size>0<\/size><\/attachment><attachment id=\"dcda8508-54f8-433b-9bfe-acddea561a27\" type=\"presentation/segment+preview\" ref=\"track:45210525-a5a8-4bab-a0a7-9451dfe8561e;time=T00:47:50:640F1000\"><mimetype>image/jpeg<\/mimetype><tags><tag>engage-download<\/tag><\/tags><url>https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/e966ff8e-ee9d-410a-bcb3-6f6f9d9de1c8/track.jpg<\/url><size>0<\/size><\/attachment><attachment id=\"2248f820-40ee-40b5-aad3-6431c963a921\" type=\"presentation/segment+preview\" ref=\"track:45210525-a5a8-4bab-a0a7-9451dfe8561e;time=T01:25:29:600F1000\"><mimetype>image/jpeg<\/mimetype><tags><tag>engage-download<\/tag><\/tags><url>https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/b812d83e-361b-4812-82cb-e092240809af/track.jpg<\/url><size>0<\/size><\/attachment><attachment id=\"570aa523-bc20-4b32-b423-903e3e149bc3\" type=\"presentation/segment+preview\" ref=\"track:45210525-a5a8-4bab-a0a7-9451dfe8561e;time=T00:12:37:160F1000\"><mimetype>image/jpeg<\/mimetype><tags><tag>engage-download<\/tag><\/tags><url>https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/3c970d74-059c-4d0a-af1d-feed5d3951f9/track.jpg<\/url><size>0<\/size><\/attachment><attachment id=\"8fad7d30-c681-47e4-bc0d-28e117fe58b0\" type=\"presentation/segment+preview\" ref=\"track:45210525-a5a8-4bab-a0a7-9451dfe8561e;time=T00:42:21:680F1000\"><mimetype>image/jpeg<\/mimetype><tags><tag>engage-download<\/tag><\/tags><url>https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/a1fe62b1-e6b6-4c49-9f7c-15131173bfac/track.jpg<\/url><size>0<\/size><\/attachment><attachment id=\"784fe4d6-c8c6-48ae-9860-adbedfafd945\" type=\"presentation/segment+preview\" ref=\"track:45210525-a5a8-4bab-a0a7-9451dfe8561e;time=T01:03:53:640F1000\"><mimetype>image/jpeg<\/mimetype><tags><tag>engage-download<\/tag><\/tags><url>https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/d5ecddad-75be-476a-9041-f408f91c7801/track.jpg<\/url><size>0<\/size><\/attachment><attachment id=\"0608259c-0d67-4022-bf57-6933b2fe870c\" type=\"presentation/segment+preview\" ref=\"track:45210525-a5a8-4bab-a0a7-9451dfe8561e;time=T00:00:42:800F1000\"><mimetype>image/jpeg<\/mimetype><tags><tag>engage-download<\/tag><\/tags><url>https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/76e6597e-d06c-462c-9370-55bbeb374af2/track.jpg<\/url><size>0<\/size><\/attachment><attachment id=\"623ba10e-6081-45d7-a8b8-74d4726f0221\" type=\"presentation/segment+preview\" ref=\"track:45210525-a5a8-4bab-a0a7-9451dfe8561e;time=T00:38:46:440F1000\"><mimetype>image/jpeg<\/mimetype><tags><tag>engage-download<\/tag><\/tags><url>https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/3166fdd2-7067-4d14-a973-178a413dd678/track.jpg<\/url><size>0<\/size><\/attachment><attachment id=\"4cc1f05f-97bc-4694-88f6-95706cf91e18\" type=\"presentation/segment+preview\" ref=\"track:45210525-a5a8-4bab-a0a7-9451dfe8561e;time=T00:40:50:440F1000\"><mimetype>image/jpeg<\/mimetype><tags><tag>engage-download<\/tag><\/tags><url>https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/72156a34-ffef-4961-9abf-8218c8d3f9f0/track.jpg<\/url><size>0<\/size><\/attachment><attachment id=\"882ae3b3-b461-4233-a9cb-09fd82856613\" type=\"presentation/segment+preview\" ref=\"track:45210525-a5a8-4bab-a0a7-9451dfe8561e;time=T01:27:46:720F1000\"><mimetype>image/jpeg<\/mimetype><tags><tag>engage-download<\/tag><\/tags><url>https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/bbb46923-b591-4330-ab8f-17a2e2d717c1/track.jpg<\/url><size>0<\/size><\/attachment><attachment id=\"a1786957-9780-4531-a73a-a4474ad1cc7b\" type=\"presentation/segment+preview\" ref=\"track:45210525-a5a8-4bab-a0a7-9451dfe8561e;time=T01:22:26:80F1000\"><mimetype>image/jpeg<\/mimetype><tags><tag>engage-download<\/tag><\/tags><url>https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/a1ae0961-a6b9-48ad-9f5f-eafce22f7656/track.jpg<\/url><size>0<\/size><\/attachment><attachment id=\"2a6a2641-e7fe-4c63-8b03-42458aa99da7\" type=\"presentation/search+preview\" ref=\"track:track-3\"><mimetype>image/jpeg<\/mimetype><tags><tag>engage-download<\/tag><\/tags><url>https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/attachment-3/track_1_000s_search.jpg<\/url><size>0<\/size><\/attachment><attachment id=\"3ee5c8ca-7909-42bd-b218-bd385285457f\" type=\"presentation/player+preview\" ref=\"track:track-3\"><mimetype>image/jpeg<\/mimetype><tags><tag>engage-download<\/tag><\/tags><url>https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/attachment-4/track_1_000s_player.jpg<\/url><size>0<\/size><\/attachment><attachment id=\"79f3e68a-2c1e-4520-ac9a-7141aaa1470a\" type=\"presentation/segment+preview\" ref=\"track:45210525-a5a8-4bab-a0a7-9451dfe8561e;time=T01:07:40:840F1000\"><mimetype>image/jpeg<\/mimetype><tags><tag>engage-download<\/tag><\/tags><url>https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/724f253a-afac-4c1b-b8c4-e01814af5a08/track.jpg<\/url><size>0<\/size><\/attachment><\/attachments><publications/><\/mediapackage>","mediaType":"AudioVisual","keywords":{"keywords":""},"modified":"2020-10-07T00:05:58+02:00","score":11.383356,"segments":{"segment":[{"index":0,"time":0,"duration":42800,"relevance":-1,"hit":false,"text":"IQ Science is to test crazy ideas Engineering is to put these ideas into Business 9HC 1 Andreas Holzinger 1 V0 709 043 Medical Informatics 30 W 30 11 2016 11 15 12 45 Lecture 06 Probabilistic Graphical 9 Models II From Bayesian Networks to 3 Graph Bandits a holzinger tugraz at Tutor markus plass student tugraz at http hci kdd org biomedical informatics big data new Group I mg 0 19 as","previews":{"preview":{"ref":"45210525-a5a8-4bab-a0a7-9451dfe8561e","$":"https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/d2128d70-db12-49ec-99e4-3f00f21cc51f/track.jpg"}}},{"index":1,"time":42800,"duration":71760,"relevance":-1,"hit":false,"text":"IU Agenda for today QHCI 06 Graph Bandits a very hot topic mz nglv Gvouv 1 709 549 as","previews":{"preview":{"ref":"45210525-a5a8-4bab-a0a7-9451dfe8561e","$":"https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/76e6597e-d06c-462c-9370-55bbeb374af2/track.jpg"}}},{"index":2,"time":114560,"duration":238600,"relevance":-1,"hit":false,"text":"TU Quiz QHCI cm l I I u N l l W n I I J E J O s 709 049 as","previews":{"preview":{"ref":"45210525-a5a8-4bab-a0a7-9451dfe8561e","$":"https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/274b400c-9e92-4b98-8b8a-52c3b852ad03/track.jpg"}}},{"index":3,"time":353160,"duration":144200,"relevance":-1,"hit":false,"text":"TU Quiz QHCI Gun l I I n M l 3 n I I J E 4 O u v i 1 u o u C 0 Nabr gImauv 5 m9 049 as","previews":{"preview":{"ref":"45210525-a5a8-4bab-a0a7-9451dfe8561e","$":"https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/d5e51fdc-4d7c-42ce-a3de-dae9ffdb8103/track.jpg"}}},{"index":4,"time":497360,"duration":58840,"relevance":-1,"hit":false,"text":"TU Quiz QHCI cm http sbcb bloch ox ac uk users ollver software Ho ur glv Gvouv e 709 5 9 as","previews":{"preview":{"ref":"45210525-a5a8-4bab-a0a7-9451dfe8561e","$":"https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/6e951bc3-5e85-4aa8-a0b2-5238edd24b13/track.jpg"}}},{"index":5,"time":556200,"duration":16720,"relevance":-1,"hit":false,"text":"Graph Mode M 01 Graphical Models and Decision laking","previews":{"preview":{"ref":"45210525-a5a8-4bab-a0a7-9451dfe8561e","$":"https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/3415228d-5667-4ed7-8379-69d95befa033/track.jpg"}}},{"index":6,"time":572920,"duration":10560,"relevance":-1,"hit":false,"text":"QHCI","previews":{"preview":{"ref":"45210525-a5a8-4bab-a0a7-9451dfe8561e","$":"https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/5891df7b-cac0-4cf9-b64f-d1049234811a/track.jpg"}}},{"index":7,"time":583480,"duration":150320,"relevance":-1,"hit":false,"text":"IU Decision Making Learn good policy for selecting actions QHCP Goal Learn an optimal policy for selecting best actions within a given context Fort 1 T 1 The world produces an uncertain context XI 6 X 2 The learner selects an action at e 1 t 3 The world reacts with a reward HUM E W Holzirglv Group 9 709 0 1905 Decision","previews":{"preview":{"ref":"45210525-a5a8-4bab-a0a7-9451dfe8561e","$":"https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/e8c5b1bf-d1de-4806-b6ac-a26482b5deac/track.jpg"}}},{"index":8,"time":733800,"duration":6480,"relevance":-1,"hit":false,"text":"QHCI","previews":{"preview":{"ref":"45210525-a5a8-4bab-a0a7-9451dfe8561e","$":"https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/15e202d1-27b0-496e-ac89-ca801fca8e2f/track.jpg"}}},{"index":9,"time":740280,"duration":16880,"relevance":-1,"hit":false,"text":"IU Key Challenges 9H0 Medicine is an extremely complex application domain dealing most of the time with uncertainties probable information When we have big data but little knowledge automatic ML can help to gain insight m 759 5 9 as","previews":{"preview":{"ref":"45210525-a5a8-4bab-a0a7-9451dfe8561e","$":"https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/15d297a9-96b7-41dd-894d-6e4fa92d1a84/track.jpg"}}},{"index":10,"time":757160,"duration":95200,"relevance":-1,"hit":false,"text":"LU Key Challenges 9H0 I Medicine is an extremely complex application domain dealing most of the time with uncertainties probable information I When we have big data but little knowledge automatic ML can help to gain insight I Structurae learning and prediction in large scale biomedical networks with probabilistic graphical models I If we have little data and deal with NP hard problems we still need the human in the loop m we 0 9 De","previews":{"preview":{"ref":"45210525-a5a8-4bab-a0a7-9451dfe8561e","$":"https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/3c970d74-059c-4d0a-af1d-feed5d3951f9/track.jpg"}}},{"index":11,"time":852360,"duration":148600,"relevance":-1,"hit":false,"text":"TU Example Why is prediction in proteins so important for us QHCI cm Noizirglv Gmuv u 709 049 De","previews":{"preview":{"ref":"45210525-a5a8-4bab-a0a7-9451dfe8561e","$":"https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/4be1e541-ff90-44b7-8f84-c116cccc87b1/track.jpg"}}},{"index":12,"time":1000960,"duration":33879,"relevance":-1,"hit":false,"text":"TU Recommended background reading cm v w Bishop C M 2007 Pattern Recognition and Machine Learnlng Heldelberg Sprunger Chapter 8 on graphical models openly available http research microsoft com en us um people cmblshop prml mmgu Group Murphy K P 2012 Machlne learnlng a probabilistic perspecllve MIT press Chapter 26 pp 907 Graphlcal model structure learnlng 11 QHCI Koller D Frledman N 2009 Probabmtlc graphical models princnples and techniques MIT press 709 549 as","previews":{"preview":{"ref":"45210525-a5a8-4bab-a0a7-9451dfe8561e","$":"https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/db18f079-68ae-447e-8e07-892c6773e8a2/track.jpg"}}},{"index":13,"time":1034839,"duration":124641,"relevance":-1,"hit":false,"text":"TU Three types of Probabilistic Graphical Models QHCI cm Undirected Markov random fields useful eg for computer vision Details Murphy 19 Directed Bayes Nets useful for designing models Details Murphy 10 y j 11 Factored useful for mlerence Iearmng I I I x II X Nolzirglr Group 13 m9 0 9 as","previews":{"preview":{"ref":"45210525-a5a8-4bab-a0a7-9451dfe8561e","$":"https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/540877d1-ab40-486f-a793-75be2b000005/track.jpg"}}},{"index":14,"time":1159480,"duration":6400,"relevance":-1,"hit":false,"text":"IU Factor Graphs learning at scaIe QHC What is the advantage of factor graphs Bayesian Networks Yes Somewhat Ancostra Gvnvrahw Process Markov Networks Yes No szm mummy mid F otm hal Factor Graphs No Yes U homt dismbuiod mforonco Tab 9 rec r ro Ra f Hemmch A hazow Noam Grove 14 m9 549 De","previews":{"preview":{"ref":"45210525-a5a8-4bab-a0a7-9451dfe8561e","$":"https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/a80a39c5-c445-4304-882f-6887423f6e96/track.jpg"}}},{"index":15,"time":1165880,"duration":48360,"relevance":-1,"hit":false,"text":"IU Factor Graphs learning at scale QHC What is the advantage of factor graphs Bayesian Networks Yes Somewhat Ancoetra Gvnvranvo Prorvss Markov Networks Yes No Lom Cont m s mm Potm hms Factor Graphs No Yes 7 an drst buif d mforonco Tab 9 rec t to RM Fero ch Amazo v mmgu Grove 4 m9 049 as","previews":{"preview":{"ref":"45210525-a5a8-4bab-a0a7-9451dfe8561e","$":"https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/ef3d6ebe-737c-42a6-aedd-9fc8ee781d11/track.jpg"}}},{"index":16,"time":1214240,"duration":41800,"relevance":-1,"hit":false,"text":"TU From structure to function prediction QHCI cm 2 0 Yowlosy 1 Slcondary semen Prodlctioll Primary an Contact and Snlvont Acctsilb lly Frodlctlon JD Struckuvo Com cl Map Baldi P Pollastrl G 2003 The prlncnpled design of large scale recursnve neural neiwork archntectures dag rnns and Ihe protein structure prediction problem The Journal of Machine Learning Research 4 575 602 Helzr gtrGrouv 15 709 my me","previews":{"preview":{"ref":"45210525-a5a8-4bab-a0a7-9451dfe8561e","$":"https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/f22280d2-d58d-4f7b-a45d-b93526d70fbe/track.jpg"}}},{"index":17,"time":1256040,"duration":62520,"relevance":-1,"hit":false,"text":"TU Protein Network Inference QHCI cm I Hypothesis most biological functions involve the interactions between many proteins and the complexity of living systems arises as a result of such interactions I In this context the problem of inferring a global protein network for a given organism I using all genomic data of the organism I is one ofthe main challenges in computational biology Yamanishi Y Vert J P Kanehisa M 2004 Protein network inference from multiple genomic data a supervised approach Biomformancs 20 Suppl 1 i363 I370 Holzir gli Group is m9 049 as","previews":{"preview":{"ref":"45210525-a5a8-4bab-a0a7-9451dfe8561e","$":"https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/a4fd407c-ab7b-4a5e-8434-7dcb1100fd5e/track.jpg"}}},{"index":18,"time":1318560,"duration":103600,"relevance":-1,"hit":false,"text":"TU Problem ls Graph lsomorphism NP complete 9H0 sun Borgwardt K M Ong C 5 Schonauer 5 1 31 s M Vishwanathan 5 Smola A J Kriegel H P A g 2005 Protein function prediction Via graph gs g 3 i O tum kernels Bioinformatics 21 Suppl 1 I47 I56 i imam mm 4min I Important for health informatics Discovering relationships between biological components I Unsolved problem in computer science I Can the graph isomorphism problem be solved in polynomial time I So far no polynomial time algorithm is known I It is also not known if it is NP complete O Holzirglv Group 7 759 5 9 as","previews":{"preview":{"ref":"45210525-a5a8-4bab-a0a7-9451dfe8561e","$":"https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/1336c585-1504-49a0-a7bf-0fc9572f8c67/track.jpg"}}},{"index":19,"time":1422160,"duration":55600,"relevance":-1,"hit":false,"text":"TU Example Protein Network Inference QHCI cm Protein network inference from multiple gag genomic data a supervised approach I I V 0 12 15 v J J NS i l lS i ch Il prrmunl Km Pl um mlcmcnnm Kg ILIK JIUJIIHIH _ KN I h lugcncllc prnlllc I Kup Kwl Km A n llnlcgmhmn nmznwcvoup 15 m9 549 De","previews":{"preview":{"ref":"45210525-a5a8-4bab-a0a7-9451dfe8561e","$":"https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/a2295e0a-f7ef-41fd-847b-40679ae28d5d/track.jpg"}}},{"index":20,"time":1477760,"duration":14760,"relevance":-1,"hit":false,"text":"TU Example Data fusion and Protein Annotation QHCI cm 3 _ r mm H m Lancknex G R De Ble T Crustlamnl N Jordan M Noble W S 2004 A statistical framework for genomic data fusnon Bloinformancs 20 16 2626 2635 Noxzmgu Gvouv 19 m9 0 1906","previews":{"preview":{"ref":"45210525-a5a8-4bab-a0a7-9451dfe8561e","$":"https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/6caf36e0-10ef-496a-adc7-2c4bc89ef2bd/track.jpg"}}},{"index":21,"time":1492520,"duration":28000,"relevance":-1,"hit":false,"text":"02 Bayesian Networks Bayes Nets","previews":{"preview":{"ref":"45210525-a5a8-4bab-a0a7-9451dfe8561e","$":"https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/f00a235f-714a-419a-b93b-f9c733c277ee/track.jpg"}}},{"index":22,"time":1520520,"duration":51480,"relevance":-1,"hit":false,"text":"TU Bayesian Network BN Definition QHCI cm I is a probabilistic model consisting of two parts I 1 a dependency structure and I 2 local probability models 1 p 1 1 n p I Pa 1 1 Where i ul arethe parents of BN nnerentlly niecel the uncertainty in the data We are a successla n arr age Demeen D csaa Iv meow and g aoh that al or to non El a 3 I dll Y e lSlD lal D osab ty C SI lDuIlO l n a sparse may on sea cn ng naepencenc elat ms in tne data Furthernrore tn 5 vrccel allows a fle ent strategies to integrate l C aata SOU CES Pearl J 1988 Probabilistic reasoning In Intelligent systems networks of plausrble Inference San Francisco Morgan Kaufmann u me 549 as um my Grove","previews":{"preview":{"ref":"45210525-a5a8-4bab-a0a7-9451dfe8561e","$":"https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/6d5a6667-c1ad-476b-9d18-56ec9de25081/track.jpg"}}},{"index":23,"time":1572000,"duration":32960,"relevance":-1,"hit":false,"text":"IU Example Directed Bayesian Network with 7 nodes 9H0 w j 2 V j 1 x v p _ _ n N V V V w V W V V V No mglv Gvouv 1 me 049 as","previews":{"preview":{"ref":"45210525-a5a8-4bab-a0a7-9451dfe8561e","$":"https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/e376250c-5db2-4426-b393-bbc4e7a2d19a/track.jpg"}}},{"index":24,"time":1604960,"duration":58360,"relevance":-1,"hit":false,"text":"TU Clinical Case Example cm gig Overmoyer B A Lee J M LerWIII M F 2011 Case 17 2011 A 49 Year Old Woman With a Mass In the Breast and Overlylng Skln Changes New England Journal of Medicine 364 23 2246 2254 Nolzwglv Gvouv","previews":{"preview":{"ref":"45210525-a5a8-4bab-a0a7-9451dfe8561e","$":"https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/519507a6-eec8-4782-b57d-9d09f351975e/track.jpg"}}},{"index":25,"time":1663320,"duration":57040,"relevance":-1,"hit":false,"text":"IU_ Important in Clinical practice prognosis QHCI the prediction ofthe future course of a disease conditional on the patient s history and a projected treatment strategy Danger probable Information Therefore valid prognostic models can be of great benefit for clinical decision making and of great value to the patient e g for notification and quality of life decisions Knaus W A Wagner D P Lynn Jr 1991 Short term mortality predictions for critically I hospitalized adults scnence and ethics Science 254 5030 389 Holzirglv Group 1 m9 0 19 as","previews":{"preview":{"ref":"45210525-a5a8-4bab-a0a7-9451dfe8561e","$":"https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/82285407-758b-480f-ae22-6a0e64912c2a/track.jpg"}}},{"index":26,"time":1720360,"duration":74560,"relevance":-1,"hit":false,"text":"TU Predicting the future on past data and present status QHCI cm t HI I CIII I ll It ll ltllt WI NINTH Itllt K RISK factors R sk factors Pathogenesns Pathogenesrs Disorders Disorders Pathophysrology Pathophysrology Flndlngs Findings Tests Treatments 0 pin ir mu past present future van Gerven MA 1 Taal B G Lucas P J F 2008 Dynamic Bayesnan networks as prognostic models for clinical patient management Journal ofBromedIcal Informatics 41 4 515 529 Narmgu Group 5 m9 5 9 as","previews":{"preview":{"ref":"45210525-a5a8-4bab-a0a7-9451dfe8561e","$":"https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/6446e2b3-f7a8-4e64-9df1-46d63281b45b/track.jpg"}}},{"index":27,"time":1794920,"duration":65560,"relevance":-1,"hit":false,"text":"TU Example Breast cancer Probability Table QHCI cm _l H w _r I r _r 1 w v mcr Wang X H et al 1999 Computer assured dragnOSIs of breast cancer usrng a data driven Bayesnan belief network International Journal of Medical Informatics 54 2 115 126 s 709 0 9 as","previews":{"preview":{"ref":"45210525-a5a8-4bab-a0a7-9451dfe8561e","$":"https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/c579ee9b-7e66-4ac9-931d-55243d9236bc/track.jpg"}}},{"index":28,"time":1860480,"duration":71800,"relevance":-1,"hit":false,"text":"TU Breast cancer big picture state of 1999 QHCI Gun Alcoholic 8 S n Smoking Thickening stchurgc Hormnncx Mcnupduu Pregnant Brcusl Cancer Ma Family Archllcclurul Ti uc Micmculu lhx lnr oDls tonion Asymmetry cutmnx Wang X H er al 1999 Computer assured dlagnOSIS of breast cancer usnng a data driven Bayesnan belief network International Journal of Medical Informatics 54 2 115 126 Nc rglvacuD 7 7m 0 1906","previews":{"preview":{"ref":"45210525-a5a8-4bab-a0a7-9451dfe8561e","$":"https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/83fb9e15-4f84-48c0-8ef9-0ac99e2300cc/track.jpg"}}},{"index":29,"time":1932280,"duration":58600,"relevance":-1,"hit":false,"text":"TU 10 years later Integration of microarray data QHCI am Integrating microarray data from multiple studies to Increase sample size approach to the development of more robust prognostic tests 3 N t I l Xu L Tan A Winslow R Geman D 2008 Merging mlcroarray data from separate breast cancer studies provndes a robust prognostic test BMC Blomformatlcs 9 1 125 139 No vaoup","previews":{"preview":{"ref":"45210525-a5a8-4bab-a0a7-9451dfe8561e","$":"https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/b29a2412-d7de-4f97-8434-0025dcb4a7a3/track.jpg"}}},{"index":30,"time":1990880,"duration":48840,"relevance":-1,"hit":false,"text":"TU Example Bayes Net with four binary variables QHCI cm Gene 1 Ptonl 0 8 Ptoffl 0 2 Gene 2 Gene 1 Gene 1 Gene 2 Gene 1 Gene 1 on 011 on 011 P1on1 O3 06 Ptonl 03 06 P101 07 04 P1011 O7 04 o Prognosns Gene 2 on Gene 2 on Gene 2 011 Gene 2 011 Gene 3 on Gene 3 of Gene 2 on Gene 3 011 Plgoodl 0 6 O 1 0 9 O 5 Plpoor O 4 O 9 O 1 O 5 Gevaen O Smet F D Tlmmerman D Moreau Y Moor B D 2006 Predicting Ihe prognosns of breast cancer by Integrating clinical and mncroarray daia wnth Bayesnan networksv Blomformancs 22 14 184 190 Ne1zr glmauD 9 m9 549 as","previews":{"preview":{"ref":"45210525-a5a8-4bab-a0a7-9451dfe8561e","$":"https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/5b258709-ba4f-46e0-aae6-49a61a64e25b/track.jpg"}}},{"index":31,"time":2039720,"duration":36320,"relevance":-1,"hit":false,"text":"TU Concept MarkowBlanket QHCI Gun Gevaert O Smet F D A Timmerman D Moreau Y Moor B D 2006 Predicting the prognosrs of breast O cancer by Integratmg J cllnlcal and mlcroarray data With Bayesran networks Blomformancs 22 14 O O 184 190 Nubr gerruup in m9 549 De","previews":{"preview":{"ref":"45210525-a5a8-4bab-a0a7-9451dfe8561e","$":"https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/39335258-3766-42e2-b74b-f21bfc4b5a12/track.jpg"}}},{"index":32,"time":2076040,"duration":24920,"relevance":-1,"hit":false,"text":"TU Dependency Structure first step 1 2 QHCI am I First the structure is learned using a search strategy I Since the number of possible structures increases super exponentially With the number of variables I the well known reed search al orithm K2 can be used In combination With the Bayesign Dirichlet BD scoring metric 1 r er r Nrk Nrk piSID xp r Ni Nij k r Ni k If I NM number of cases in the data set D havmg variable 1 in state associated With the i th instantiation of its parents in current structure 5 n is the total number of variables Holzr gerrouD 31 me 0490","previews":{"preview":{"ref":"45210525-a5a8-4bab-a0a7-9451dfe8561e","$":"https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/be112ef5-6009-4bc3-a476-4ecb3a9d77a0/track.jpg"}}},{"index":33,"time":2100960,"duration":7360,"relevance":-1,"hit":false,"text":"TU Dependency Structure first step 2 2 QHCI cm I Next rs calcu ated by summing over aH states of a varrabte I t _ t _ and have snmr ar meanrngs but refer to prror 0 knowledge for the parameters I When no khowtedge IS avarlab e they are estrmated usmg X r r I t I wrth Vthe equrva entsamp e srze I r thehumber of states ofvarrab ez and I L the number of mstantratrons of the parents of varrablez I It Icorrespondstothe gamma distribution I FmaHy pm t Is the prlor probablhty ot the structure I mMrs catcu ated by NW I II 11 I rll 1I 1 J I wrth 1 the number of parents of varlabte and u aH the varrables that are not a parent of x I Next pm In Is the probabrhty that there IS an edge from u to 7 whr e rm r rs the mverse e the probabrhty that there 15 no edge from u to Ir narmgu Grove a1 m9 5 9 as","previews":{"preview":{"ref":"45210525-a5a8-4bab-a0a7-9451dfe8561e","$":"https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/ea79344c-865a-4898-970a-5f1caead5f7a/track.jpg"}}},{"index":34,"time":2108320,"duration":43680,"relevance":-1,"hit":false,"text":"TU Parameter learning second step QHCI Gun Estlmatlhg the parameters of the local probablllty models correspondlng Wlth the dependency structure CPTs are used to model these local probablllty models For each varlable and mstantlatlon of lts parents there exlsts a CPT that conslsts of a set of parameters Each set of parameters was glven a uniform Dirichlet prlor 11 HH S DH LHHIN HI WM WWIZA VJH W l 7 cefmec as before 709 0 9 as Nol glv Grove 33","previews":{"preview":{"ref":"45210525-a5a8-4bab-a0a7-9451dfe8561e","$":"https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/f1acca10-7339-48f3-b8a8-45d235c6b0c9/track.jpg"}}},{"index":35,"time":2152000,"duration":52280,"relevance":-1,"hit":false,"text":"TU Predicting the prognosis of breast cancer integrated a QHCI cm mpmr H mv l Iv lunmrmmv l1 N Ulml V K Mm n I wnnml v wnuug T rp ngl u l VIIII JWI m u mm L4 um m1 mm um m mum w 2 4 u My m9 0 9 as u Hoh r glv Grove","previews":{"preview":{"ref":"45210525-a5a8-4bab-a0a7-9451dfe8561e","$":"https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/f9ba87ff-9ee0-4adc-a383-1b00646b7527/track.jpg"}}},{"index":36,"time":2204280,"duration":122160,"relevance":-1,"hit":false,"text":"LU Inference in Bayes Nets is intractable NP complete 9H0 I For certain cases it is tractable if I Just one variable is unobserved I We have singly connected graphs no undirected loops belief propagation I Assigning probability to fully observed set of variables I Possibility Monte Carlo Methods generate many samples according to the Bayes Net distribution and then count the results I Otherwise approximate solutions NOTE Sometimes it is better to have an approximate solution to a complex problem than a perfect solution to a simplified problem Nolzr gerrouD m9 5 19 as","previews":{"preview":{"ref":"45210525-a5a8-4bab-a0a7-9451dfe8561e","$":"https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/91059181-d401-4617-b548-81e0723871da/track.jpg"}}},{"index":37,"time":2326440,"duration":7480,"relevance":-1,"hit":false,"text":"3 Machine Learning on Graphs","previews":{"preview":{"ref":"45210525-a5a8-4bab-a0a7-9451dfe8561e","$":"https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/3166fdd2-7067-4d14-a973-178a413dd678/track.jpg"}}},{"index":38,"time":2333920,"duration":37360,"relevance":-1,"hit":false,"text":"I U_ Example Lymphoma is the most common blood cancer 9H0 The two main forms of lymphoma are Hodgkin lymphoma and non Hodgkin lymphoma NHL Lymphoma occurs when cells of the immune system called lymphocytes a type of white blood cell grow and multiply uncontrollably Cancerous lymphocytes can travel to many parts of the body including the lymph nodes spleen bone marrow blood or other organs and form a mass called a tumor The body has two main types of lymphocytes that can develop into lymphomas B lymphocytes B cells and T lymphocytes T cells www lymphomaorg http imagebankhematologyorg Holzir gtv Swan 37 709 0 19 06","previews":{"preview":{"ref":"45210525-a5a8-4bab-a0a7-9451dfe8561e","$":"https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/9af7c54f-3da7-49c3-852e-bf3b66f6aded/track.jpg"}}},{"index":39,"time":2371280,"duration":79160,"relevance":-1,"hit":false,"text":"TU ML tasks on graphs QHCI Gun Discover unexplored interactions in PPI networks and gene regulatory networks Learn the structure Reconstruct the structure _ 7 1 Dlnrvch M T Klau G W Rosenwald A Dandekar T Muller T 2008 Idenhfymg functional modules In protein protein lnteractxon networks an Integrated exact approach Blomformatlcs 24 13 l223 l231 Holzlrglr Gvouu BS 709 5 9 as","previews":{"preview":{"ref":"45210525-a5a8-4bab-a0a7-9451dfe8561e","$":"https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/9ccda672-3ca8-40f7-a797-86c4ccc1daf6/track.jpg"}}},{"index":40,"time":2450440,"duration":23880,"relevance":-1,"hit":false,"text":"TU From structure to function QHCI cm _ _ love I J A Protocol for Computer BJsed Protein Structure and runcnon Predmmn mun 0 _u EEK I lvv CKAIVY C V Umzi http www10ve com video 3259 a protocoHor computer based proteln structure tunctvon new 5mm 39 me 049 as","previews":{"preview":{"ref":"45210525-a5a8-4bab-a0a7-9451dfe8561e","$":"https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/72156a34-ffef-4961-9abf-8218c8d3f9f0/track.jpg"}}},{"index":41,"time":2474320,"duration":67360,"relevance":-1,"hit":false,"text":"TU Interesting Hubs tend to link to small degree nodes QHCI cm Nodes proteins Links physical interactions binding Puzzling pattern Hubs tend to link to small degree nodes Why IS this puzzling in a randcm r et xcrk the meta it ma 3 noce x 1 1 degree l 5 to a nose nit egree lS AA I p 015 p 0001 Jeong H Mason S P Barabasn A L Oltvai Z N 2001 Lethality and centrality In protein networks Nature 411 6833 41 42 709 049 as AU mimgu Grouv","previews":{"preview":{"ref":"45210525-a5a8-4bab-a0a7-9451dfe8561e","$":"https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/a0c59d71-959f-4c49-ab51-5e462bf4c221/track.jpg"}}},{"index":42,"time":2541680,"duration":9680,"relevance":-1,"hit":false,"text":"TU Interesting Hubs tend to link to small degree nodes QHCI cm Nodes proteins Links physical interactions binding Puzzling pattern Hubs tend to link to small degree nodes Why IS this puzzling in a anacm net usrh tr e Drcbab it ma a nod eg ee l x510 a node x 1 egree ls i M I I 2 3 453 25 p l 1 p 0 mm Jeong H Mason S P Barabasn A L Oltvai Z N 2001 Lethality and centrality In protein networks Nature 411 6833 41 42 mg 0 1un a elm 5mm","previews":{"preview":{"ref":"45210525-a5a8-4bab-a0a7-9451dfe8561e","$":"https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/a1fe62b1-e6b6-4c49-9f7c-15131173bfac/track.jpg"}}},{"index":43,"time":2551360,"duration":39560,"relevance":-1,"hit":false,"text":"TU Example Subgraph Discovery QHCI cm MIMIJI V K 4 l4 I M Um z ml WWW r w w it w s_ n ppm huki twqr a 1 NR HGQ JM MW mum wmvcyxw W 1111 NM HRH H ul l Nehr glvaouD 41 m9 5 905","previews":{"preview":{"ref":"45210525-a5a8-4bab-a0a7-9451dfe8561e","$":"https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/f0952a7a-d650-4656-90f4-bb7e7dc5e949/track.jpg"}}},{"index":44,"time":2590920,"duration":5280,"relevance":-1,"hit":false,"text":"TU Example Subgraph Discovery QHCI cm do saw m h String My HIGH ENERGY PHYS CS THEORY ASTROPHYSICS Hm M W W Model 01 um Anbotfopy MMNMM An Alternative To Compacmication lesnmaled bndgeness 1276 GENERAL RELAYIVITY AND QUANTUM COSMOLOGY K x uh n M mu um mm lwau mvlJppwg A mum mu mmm AWMHW l vlsmvkx Lue vml gx a n maSnnIExv Dimnuon HIGH ENERGY PHYS CS Ndmml m hh lll 4 m mu v x PHENOMENOLOGV 11th l in 14 H m mums Gvoup u 709 0 1905","previews":{"preview":{"ref":"45210525-a5a8-4bab-a0a7-9451dfe8561e","$":"https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/e9325995-b460-4c65-9c8a-72f0e3972b25/track.jpg"}}},{"index":45,"time":2596200,"duration":152200,"relevance":-1,"hit":false,"text":"TU Why do we want to apply ML to graphs QHCI cm I A Discovery of unexplored interactions I 8 Learning and Predicting the structure I C Reconstructing the structure I Which joint probability distributions does a graphical model represent I How can we learn the parameters and structure of a graphical model u m w m The chemical space O possmle small or ganic molecules 3 w stars In the observ able universe Nolzmglv 6mm 41 m9 on as","previews":{"preview":{"ref":"45210525-a5a8-4bab-a0a7-9451dfe8561e","$":"https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/e62c918e-485c-45c6-9e7f-742c7e1446c4/track.jpg"}}},{"index":46,"time":2748400,"duration":39920,"relevance":-1,"hit":false,"text":"TU Example Question Predicting Function from Structure QHCI cm 11 my I I vasmwuuvxm magnum mm um I vasmmvxm magnum mam um I SA vaounuLI1I Hal 4 VIVIXGKLDILI It xm v I am x I mun cu r I m1 x n not n x v V I nu max I Inuuuru mm I m1 I u 19me I n 2a 2 O n z 4 How similar are two graphs How similar is their structure How similar are their node and edge labels LHKLl M KAI hill A l 1mm ll Kn ll lllp i Lil Illtn i v l l H Hi llllh Jr hmi l 1 Vi i vl lr m mtunui mm i Ilhi rn mitrawmmwy nu ms nu I m Noizr glmauD 13 m9 549","previews":{"preview":{"ref":"45210525-a5a8-4bab-a0a7-9451dfe8561e","$":"https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/6ec9c41f-f1dd-41d8-966c-cc852d005075/track.jpg"}}},{"index":47,"time":2788320,"duration":82320,"relevance":-1,"hit":false,"text":"TU Graph Comparison QHCI cm I Similar Property Principle Molecules having similar structures should have similar activities I Structure based representations Compare molecules by comparing substructures e g I Sets as vectors Measure similarity by the cosine distance I Sets as sets Measure similarity by the Jaccard distance I Sets as points Measure similarity by Euclidean distance I Problems Dimensionality Non Euclidean cases Noimgu Group 4 m9 049 as","previews":{"preview":{"ref":"45210525-a5a8-4bab-a0a7-9451dfe8561e","$":"https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/ffac8015-7952-4061-a9c5-00c95b5de942/track.jpg"}}},{"index":48,"time":2870640,"duration":16880,"relevance":-1,"hit":false,"text":"4 Little Excursus What is similarity","previews":{"preview":{"ref":"45210525-a5a8-4bab-a0a7-9451dfe8561e","$":"https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/e966ff8e-ee9d-410a-bcb3-6f6f9d9de1c8/track.jpg"}}},{"index":49,"time":2887520,"duration":56160,"relevance":-1,"hit":false,"text":"TU What is Similar QHCI cm Image credlt tooEamonn Keogh 2008 Ne pr glv Gvouo 16 709 on as","previews":{"preview":{"ref":"45210525-a5a8-4bab-a0a7-9451dfe8561e","$":"https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/dffc5f4a-3b7d-4af9-9753-04151aaca684/track.jpg"}}},{"index":50,"time":2943680,"duration":55440,"relevance":-1,"hit":false,"text":"TU QHCI Gru i M m I M M M mu 4 N wm rn 1 41er nxlwm mm mwm t mmw Noungucveup 47 m9 0 as","previews":{"preview":{"ref":"45210525-a5a8-4bab-a0a7-9451dfe8561e","$":"https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/6dfc7169-200b-4084-b780-bfd15c6e3651/track.jpg"}}},{"index":51,"time":2999120,"duration":18360,"relevance":-1,"hit":false,"text":"TU QHCI cm M vam A M 4 IHIH M 11 l M 5 w 0 Mum m u h wrm U mm mm mum Mm x t mmw vaouD 4s me 5 9 as","previews":{"preview":{"ref":"45210525-a5a8-4bab-a0a7-9451dfe8561e","$":"https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/c342a425-cabb-40a0-a732-535300542d22/track.jpg"}}},{"index":52,"time":3017480,"duration":88000,"relevance":-1,"hit":false,"text":"TU Similarity and Correspondence QHCI cm lnn V l A J W H I l J M li i i JW X NL HWILJI rmvmvr Mum mmwumx Mun me mr e http www lnf usi ch bronstein Structure Structure Neizr glvaouD 49 m9 549 as","previews":{"preview":{"ref":"45210525-a5a8-4bab-a0a7-9451dfe8561e","$":"https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/9724eb4c-9371-41a2-84b0-b216a65730fa/track.jpg"}}},{"index":53,"time":3105480,"duration":19400,"relevance":-1,"hit":false,"text":"TU Invariant Similarity QHCI cm Transformanon naimgu Gvouv so 709 049 as","previews":{"preview":{"ref":"45210525-a5a8-4bab-a0a7-9451dfe8561e","$":"https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/53f08c19-ebc5-4ce7-9ddb-2681eb42c00c/track.jpg"}}},{"index":54,"time":3124880,"duration":87320,"relevance":-1,"hit":false,"text":"I Gromov Hausdorff dist finding the opt correspondence Gromov M 1984 infinite groups as geometric objects Felix Hausdorff Michall Gromov 1868 1942 1943 w Ym y Metric space Metric space 1 IGM V Y min max r r n 1 u g i _ i t r lh S tr r LY I h Li S t I 2 L Discrete optimization over correspondences is NP hard Neimgu Group 51 m9 0 1906","previews":{"preview":{"ref":"45210525-a5a8-4bab-a0a7-9451dfe8561e","$":"https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/d6759621-7816-423d-9969-dd4cccb1b4ba/track.jpg"}}},{"index":55,"time":3212200,"duration":70480,"relevance":-1,"hit":false,"text":"TU Example QHCI cm 0 Sormanv C 2010 How Riemanman Manlfokis Converge A Survey arXIv preprml awn 1006704117 mums Gvouv 5 me my 05","previews":{"preview":{"ref":"45210525-a5a8-4bab-a0a7-9451dfe8561e","$":"https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/8c57954f-4451-45f0-98ba-131bb2f55272/track.jpg"}}},{"index":56,"time":3282680,"duration":9280,"relevance":-1,"hit":false,"text":"5 Probabilistic Topic Models","previews":{"preview":{"ref":"45210525-a5a8-4bab-a0a7-9451dfe8561e","$":"https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/5f67c6bd-1941-4052-a696-96aa17daefbc/track.jpg"}}},{"index":57,"time":3291960,"duration":102080,"relevance":-1,"hit":false,"text":"TU Topic modelling small topic but hot topic in ML QHCI cm TOPIC MODELING PROBABILISTIC MODELING STATISTICS MACHINE LEARNING DATA SCIENCE mg 549 De Nolzwglv Gvouv 54","previews":{"preview":{"ref":"45210525-a5a8-4bab-a0a7-9451dfe8561e","$":"https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/35a92b3f-2e1e-4308-9c90-0a2b39f6bb51/track.jpg"}}},{"index":58,"time":3394040,"duration":57880,"relevance":-1,"hit":false,"text":"TU Geometry of Topic Models QHCI cm I l urd I Q loplc O Uh cn cd documcnl 8 gcncrulcd ducumcnl I mrdlh I ord Blei D M 2012 Probabilistic topic models Communications of the ACM 55 4 77 84 d0I 10 1145 2133806 2133826 Noungucveup 55 m9 049 De","previews":{"preview":{"ref":"45210525-a5a8-4bab-a0a7-9451dfe8561e","$":"https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/eee918ba-be51-4bd6-97a0-ab7f740bd4c7/track.jpg"}}},{"index":59,"time":3451920,"duration":131960,"relevance":-1,"hit":false,"text":"TU Generative statistical model for natural language QHCI cm Seelung Lile s Bare Genetic Necessities H__ J Given the parameters a and E the joint distribution of a topic mixture 9 a set of N topics Z and a set of N words w is given by 9 L 1 15 M Win i311i9 Wu 3 III 0 Blei D M Ng A Y Jordan M 2003 Latent dirichlet allocation The Journal of machine Learning research JMLR 3 993 1022 Hoizr guGrouD 55 m9 0 1906","previews":{"preview":{"ref":"45210525-a5a8-4bab-a0a7-9451dfe8561e","$":"https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/57a64a0d-b0dc-4640-8237-a4df5ef3f941/track.jpg"}}},{"index":60,"time":3583880,"duration":102680,"relevance":-1,"hit":false,"text":"TU Motivation to get insight into unknown document sets QHCI cm http ago dst glthub Io dfr browser demo N modeI scaled Noizr glvaouD 57 m9 M906","previews":{"preview":{"ref":"45210525-a5a8-4bab-a0a7-9451dfe8561e","$":"https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/de1e857b-3297-4b8e-8197-4734ae81cf41/track.jpg"}}},{"index":61,"time":3686560,"duration":71400,"relevance":-1,"hit":false,"text":"TU Example from Bioinformatics QHCI cm A Funcmrul moon 1 Fur mom mod m 2 arm Emm Emmv 2 Er zr w 3 Emm 4 1 Ga J I l Em am m 3 I Yogic 1 Yo 2 P nlmr 1 WW Llnmbom pairg O B mama mmm 9mm C sum Imam n Konieuny S G Diexz L Mchardy A C 2011 Inferring functional modules of proteln famllles wnth probabilistic oplc models BMC blonnformancs 12 1 1 Nmzngusmup 55 mg 0 19","previews":{"preview":{"ref":"45210525-a5a8-4bab-a0a7-9451dfe8561e","$":"https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/661b3521-6399-4131-b8d2-bfd5c1d47441/track.jpg"}}},{"index":62,"time":3757960,"duration":27640,"relevance":-1,"hit":false,"text":"TU Eval scheme for inferred potential functional modules QHCF cm 06 am an 064 an ac ac on m 4 0 1 on or Hg 0 on ocv x wan y H I J x MJuImy A x 1m Ir m v g mn mun mum w vlrw amulm mv wuuumn m mum qu muu mum 2 m 1 Nc rglvaouD 59 m9 5 905","previews":{"preview":{"ref":"45210525-a5a8-4bab-a0a7-9451dfe8561e","$":"https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/17e75605-f22c-49d6-9fbc-c0f4900bcfbc/track.jpg"}}},{"index":63,"time":3785600,"duration":48040,"relevance":-1,"hit":false,"text":"TU Example QHCI am Toplc proportions and T m n epics Docu e 5 ass gnmenls Seeking Life s Bare Genetic Necessities 2 We only observe the docs the other structure is hidden then we compute the posterior p t p a docs Hcizrg vaouv so my 049 De","previews":{"preview":{"ref":"45210525-a5a8-4bab-a0a7-9451dfe8561e","$":"https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/82b5188e-6bfa-463f-a815-e6bfc091c6eb/track.jpg"}}},{"index":64,"time":3833640,"duration":34960,"relevance":-1,"hit":false,"text":"IU_ We find topics through a Generative Probabilistic Model 9H0 Goal to get insight in unknown document collections See a nice demo http agoldst github Io dfr browser demo Wmodel gnd Topic pmpomons and TODICS Documenls asstgnmenls Seeking Live s Bare Genetic Necessities V z X gt 3 5 i Each doc is a random mix of corpus wide topics and each word is drawn from one of these topics Noimgu Giana si 759 on De","previews":{"preview":{"ref":"45210525-a5a8-4bab-a0a7-9451dfe8561e","$":"https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/d5ecddad-75be-476a-9041-f408f91c7801/track.jpg"}}},{"index":65,"time":3868600,"duration":8800,"relevance":-1,"hit":false,"text":"TU Output Example 4 learned topics details in Blei 2008 QHCI cm llumnn mnlnlinn llM ilM Hlllllllll l gl llnnn vmlutimuu y In Illml l Inn Ill l Inll ll l lzl inl nl nmliull gnu lil ul gnlli lls mum s lulu gt lll s lll t l l si zlllt v ulll llll l M qlu nu origin Inll ll l lill _ ll lll 34mm llll nluuy 111 1le u k Illnlwlllnr grmu lmiI1 y lvm Mum min pln lngvnvlix rumml lIllIlll l map living inlwtirgb pnmlll l inlurnmliun xli 1 il malaria llll l lml gt m Iim gmup I ll l ll Ilt l l k mapping nvw ll il lll wl lwnn prnjm t l u llllllHl 11 wqut lnw Hllllllnll llllll l l lll l l simululiuns Hal glv Evans 6 m9 L349 05","previews":{"preview":{"ref":"45210525-a5a8-4bab-a0a7-9451dfe8561e","$":"https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/f0de732e-5ddd-4caf-ab64-ba2e7939cb70/track.jpg"}}},{"index":66,"time":3877400,"duration":17880,"relevance":-1,"hit":false,"text":"TU Output Example 4 learned topics details in Blei 2008 QHCI cm llumun mnlnlinn lllN ilN unnpnlvl gummv ulutinnulj llmt Inmlvl Inn Ill l lml ll l lzl inllu umliull ngIl lil ul gilllisllls mum s lulu gt llm lll t l l sl zllll v llllllllll l h N qllt nu ul iglll Inll ll l lill _ l lll gvnv llll Ill my 111 1le 1 k Illnlm lllnr grmq ll illlls x slvnh l quvnri11g pln lngvnvlit t nllll nl llllllll l map living illl wtium pnrnllvl inlnrmnliun xli r il Illulul izl lnvlllmls ut m I i gm mp l ll 1 il m tm rk Innppillu nvw lmrmitlw wl nmn j l l u llllilwl 11 N qllt llt l s l HllllllHll llllN l l lllH l simululimls mum Evans 51 m9 L349 us","previews":{"preview":{"ref":"45210525-a5a8-4bab-a0a7-9451dfe8561e","$":"https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/addb0b35-fee3-4ef9-9680-e9d913c71982/track.jpg"}}},{"index":67,"time":3895280,"duration":13600,"relevance":-1,"hit":false,"text":"TU LDA is an example for a probabilistic graphical model QHCI cm V v V V v s a I no L n H r p i I K Encodes assumptions on data With a factorization of the Jomt Connects assumptions to algorithms for computing With data Defines the posterior through the pint HolzngliGiouD 53 m9 04905","previews":{"preview":{"ref":"45210525-a5a8-4bab-a0a7-9451dfe8561e","$":"https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/ca1ad350-3192-41f2-b2be-734b1bba7373/track.jpg"}}},{"index":68,"time":3908880,"duration":40000,"relevance":-1,"hit":false,"text":"TU LDA is an example for a probabilistic graphical model QHCI cm u it in n H u p i I K Encodes assumptions on data With a factorization of the iomt Connects assumptions to algorithms for computing With data Defines the posterior through the pint Nolzr gliGiouD 53 709 54905","previews":{"preview":{"ref":"45210525-a5a8-4bab-a0a7-9451dfe8561e","$":"https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/835ca2af-7235-4911-aa10-c3442fcbf6a6/track.jpg"}}},{"index":69,"time":3948880,"duration":20480,"relevance":-1,"hit":false,"text":"IU Posterior inference QHCI I H l n Ill g 7 ri 0 z w l 0 it l 0 z w We can t compute the denominator the marginal p w therefore we use approximate inference h 0 zl However this do not scale well Noimgu new M m9 549 as","previews":{"preview":{"ref":"45210525-a5a8-4bab-a0a7-9451dfe8561e","$":"https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/9c3c7f8a-c597-4d07-bc3b-62bae8b79223/track.jpg"}}},{"index":70,"time":3969360,"duration":42480,"relevance":-1,"hit":false,"text":"TU For big data stochastic variational inference QHCI cm Hoffman M D Blei 0 M Wang C Paislev J 2013 Stochastic variational inference The Journal of Machine Learning Research 14 1 1303 1347 NeizrglvaouD as 709 on us","previews":{"preview":{"ref":"45210525-a5a8-4bab-a0a7-9451dfe8561e","$":"https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/33c4b33a-8a0a-45f8-9086-354b82a36b26/track.jpg"}}},{"index":71,"time":4011840,"duration":23480,"relevance":-1,"hit":false,"text":"IU Stochastic variational inference in LDA QHCI 44 Wu H v v v Li H U Ji U z r DIR UV I A o i Sammie a oocumem Estimaie me ocai war a O ai parawete s JS 9 We CLANOHI topics FO V HWIO WCO GTO ICO CS VO W TT OSC OCH oa av eters poem Upcate topics as a eg 10d average 0 wormed me are current Ionics NuiznglvaouD as m9 549 De","previews":{"preview":{"ref":"45210525-a5a8-4bab-a0a7-9451dfe8561e","$":"https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/68c1787c-db47-4066-80ee-185a2205586c/track.jpg"}}},{"index":72,"time":4035320,"duration":8360,"relevance":-1,"hit":false,"text":"TU For big data stochastic variational inference QHCI cm Hoffman M D Blei Dr M Wang C Paislev J 2013 Stochastic variational inference The Journal of Machine Learning Research 14 1 1303 13477 Ncizr gquvoup es 709 0 9 as","previews":{"preview":{"ref":"45210525-a5a8-4bab-a0a7-9451dfe8561e","$":"https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/9abd0d03-0891-458b-8006-0a911695502d/track.jpg"}}},{"index":73,"time":4043680,"duration":17160,"relevance":-1,"hit":false,"text":"IU Stochastic variational inference in LDA QHCI u if H v v v Li H izm a r 3 U I A O i Samo e a Document ESIHWBIC 10 OCQ JEN 21 mai C8 Z J C C S iSi 10 CLM CHI 00 CS Pow interrveo ate Ionics rO n r ose ioca Caraweiers me Upcate topics as a we gr ted average 0 mowed are are curren 10 cs Noizr glvaouD as 7591 4905","previews":{"preview":{"ref":"45210525-a5a8-4bab-a0a7-9451dfe8561e","$":"https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/1e052c0a-5870-4dbe-ad1b-3c15715ccbd2/track.jpg"}}},{"index":74,"time":4060840,"duration":27080,"relevance":-1,"hit":false,"text":"TU Stochastic variational inference QHCI cm lmiuiil andumh cl m iCP xll hcdulc lppluprmlk h rvpul nnpic i dmumcni L umiumih Imxn Ihc LILI Ci lmluii c g 1 In v 1 A rvpul M X I J _ unlil liui mmxncmx i JUL 4 wnxnuc m 1 A 1 cl mmnmimc Alj hx i 1r l 17 i A unlil lurcwi Hoffman M D Blei D M Wang C Paislev J 2013 Stochastic variational inference The Journal of Machine Learning Research 14 1 1303 1347 vamm 57 ms 049 as","previews":{"preview":{"ref":"45210525-a5a8-4bab-a0a7-9451dfe8561e","$":"https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/724f253a-afac-4c1b-b8c4-e01814af5a08/track.jpg"}}},{"index":75,"time":4087920,"duration":35200,"relevance":-1,"hit":false,"text":"IU Approximate inference can be difficult to achieve QHCI KNOWLEDGE v v Make assumptions Discover patterns Predict 81 Explore Q 1 I tn Approximate inference can be difficult to derive Especially true for models that are not conditionally conjugate Discrete choice modelst Bayesian generalized linear models Holds us back from trying many models Holzr gerrouD as me 5 1906","previews":{"preview":{"ref":"45210525-a5a8-4bab-a0a7-9451dfe8561e","$":"https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/f16d5787-f7b0-4635-8675-8ea5c56411c1/track.jpg"}}},{"index":76,"time":4123120,"duration":37679,"relevance":-1,"hit":false,"text":"TU Black Box Approach QHCI cm REUSABLE MASS VE VARIATIONAL DATA FAMILIES ANY MODEL BLACK BOX _ VARIATIONAL INFERENCE O Easily use variational inference with any model No exponential family requirements No mathematical work beyond specifying the model Nolzr glmauD 59 m9 5 9 as","previews":{"preview":{"ref":"45210525-a5a8-4bab-a0a7-9451dfe8561e","$":"https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/13803432-0117-4425-9cca-21ef610c3271/track.jpg"}}},{"index":77,"time":4160799,"duration":164201,"relevance":-1,"hit":false,"text":"IU_ Conclusion What are future challenges 9 Flexible and expressive components for building models are of utmost importance Scalable and generic inference algorithms multi task and transfer learning Usability gets a totally new importance Easy to use algorithms for the non expert user to stretch probabilistic modeling into new areas Topic models are one approach towards detection of topics in document collections I Example Identifying re occurring patterns in such data collections gaining new knowledge Nolzr gerrouD 70 m9 mg as","previews":{"preview":{"ref":"45210525-a5a8-4bab-a0a7-9451dfe8561e","$":"https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/4cef5745-e19a-4571-9b11-84b3948d8d8e/track.jpg"}}},{"index":78,"time":4325000,"duration":45680,"relevance":-1,"hit":false,"text":"6 Graph Bandits","previews":{"preview":{"ref":"45210525-a5a8-4bab-a0a7-9451dfe8561e","$":"https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/d7ec3d3e-0242-4bb6-ae8e-10e8cc9d8b47/track.jpg"}}},{"index":79,"time":4370680,"duration":29720,"relevance":-1,"hit":false,"text":"TU The complexities of optimization S basitien Bubeck QHCI am l m a bandit Bubeck S Cesa Bianchi N 2012 Regret Analysns of Stochastic and ORF523 The complexmes of Nonstochastic Multi _ _ armed Bandit optimization Problems Machine Learning 5 1 1 122 hnps blogs pnncetonedu imabandit Also very Interesting Bubeck S 2015 Convex optuzanon Algorithms and complexity Foundations and Trends in Machine Learning 8 134 231 357 Newgu 6mm 7 759 on us","previews":{"preview":{"ref":"45210525-a5a8-4bab-a0a7-9451dfe8561e","$":"https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/cd276b03-f8c6-443b-b416-4d4ed7385e1b/track.jpg"}}},{"index":80,"time":4400400,"duration":40200,"relevance":-1,"hit":false,"text":"6 Graph Bandits","previews":{"preview":{"ref":"45210525-a5a8-4bab-a0a7-9451dfe8561e","$":"https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/63a39edf-fc5e-43b7-8b22-dad80bdda431/track.jpg"}}},{"index":81,"time":4440600,"duration":253640,"relevance":-1,"hit":false,"text":"IU What is a bandit 9 I Slot machine bandit robs your money I One armed bandit I Very simple model for sequential decision making under uncertainty I Main challenge exploration versus exploitation I Many application domains A B Testing Crowdsourcing optimization search Nclzmgu Grouv 73 m9 049 as","previews":{"preview":{"ref":"45210525-a5a8-4bab-a0a7-9451dfe8561e","$":"https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/e2acd983-4a95-424d-8ad9-73111a1d8178/track.jpg"}}},{"index":82,"time":4694240,"duration":126800,"relevance":-1,"hit":false,"text":"TU Multi Armed Bandits problem QHCI om I Multi armed bandit a gambler strategically operating multiple machines in order to draw the highest possible profits I There are n slot machines einarmige Banditen I Each machine 1 returns a reward y z P y 9 I Challenge The machine parameter 0 is unknown I Which arm of a slot machine should a gambler pull to maximize his cumulative reward over a sequence of trials stochastic setting or adversarial setting Nolzr glmauv 7 759 549 as","previews":{"preview":{"ref":"45210525-a5a8-4bab-a0a7-9451dfe8561e","$":"https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/3016dfa2-6f63-4838-905a-56503c1ddea3/track.jpg"}}},{"index":83,"time":4821040,"duration":58200,"relevance":-1,"hit":false,"text":"IU Machine Parameters of the k armed Bandit QHCI Each arm a either wins reward 1 with fixed unknown probability 1 or loses reward 0 with fixed unknown probability 1 1 All draws are independent given ill m Problem How to pull arms to maximize the total reward O Nclzir gli Evans 75 m9 049 as","previews":{"preview":{"ref":"45210525-a5a8-4bab-a0a7-9451dfe8561e","$":"https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/98f14e9e-0bd5-4814-accc-27dd0c253161/track.jpg"}}},{"index":84,"time":4879240,"duration":66840,"relevance":-1,"hit":false,"text":"TU Underlying Principle of the k Armed Bandits problem 9H0 sun I Let 1 E l n be the cltoice of a machine at time r I Let y E E be the outcome With a mean of hi I Now the given policy maps all history to a new chOIce I i i uh p ii I The problem Find a policy n that m iX y I Now two effects appear when choosing such machine I You collect more data about the machine knowledge I You collect reward I Exploration and Exploitation I Exploration Choose the next action Ll to miiiilli Ii H I Exploitation Choose the next action it to iiiuxiv I models an agent that simultaneously attempts to acquire new knowledge called exploration and optimize his or her deCISIons based on eXIsting knowledge called eprOitation The agent attempts to balance these competing tasks in order to maximize total value over the period of time considered More information http researchmicrosoft com en us proiects bandits Nolzr glvaouD 75 m9 0 9 as","previews":{"preview":{"ref":"45210525-a5a8-4bab-a0a7-9451dfe8561e","$":"https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/2d278a9a-1a23-449f-8cdd-718ef36b6bf5/track.jpg"}}},{"index":85,"time":4946080,"duration":99400,"relevance":-1,"hit":false,"text":"TU MAP Principle Optimism in the face of uncertainty QHCI cm 1 Itm xtlwv M 11 1HH M _ EM the higher the estimated the higher the W reward the higher the chance uncertainty the higher the to select the action chance 0 select the action Auer P Cesa Blanchl N Fischer P 2002 Flnlte time analysns of the multlarmed bandit problem Machlne learning 47 2 3 235 256 Nob mglvaouD 77 759 049 as","previews":{"preview":{"ref":"45210525-a5a8-4bab-a0a7-9451dfe8561e","$":"https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/a1ae0961-a6b9-48ad-9f5f-eafce22f7656/track.jpg"}}},{"index":86,"time":5045480,"duration":23400,"relevance":-1,"hit":false,"text":"TU A bandit in a graph is still a bandit QHCI cm 0 Let G a mown graph With K nodes 11 2 K Lc t f be 3 unknown function defined on the set 0 nodes oFort lton 5 w 1 may Obsuu usuard i 1 o 603 mmimizc sum of apt ted rewards Equwaientiv minimize regret R ir r u 1 where 7 mm A Mi 0 C Care about the Case when H n Ncizmglv Gvouo 73 ms 549 De","previews":{"preview":{"ref":"45210525-a5a8-4bab-a0a7-9451dfe8561e","$":"https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/1b3fc3a1-9d79-42f4-9d9c-9809b2121efa/track.jpg"}}},{"index":87,"time":5068880,"duration":24200,"relevance":-1,"hit":false,"text":"TU Smooth Graph Function QHCI cm Nobr glvaouD 79 759 my me","previews":{"preview":{"ref":"45210525-a5a8-4bab-a0a7-9451dfe8561e","$":"https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/628b2a11-514b-4fa3-8ba9-9aca78ad4e2e/track.jpg"}}},{"index":88,"time":5093080,"duration":6560,"relevance":-1,"hit":false,"text":"TU Knowledge Representation in MAB QHCI cm Knowledge can be represented in two ways l as fullhistory f l l 1 7 V 1 or 2 as belief 1 Hi H I where 0 are the unknown parameters c6 all machines The process can be modelled as belief MDP 1H otherwnse an 709 my me Nol 3 Group","previews":{"preview":{"ref":"45210525-a5a8-4bab-a0a7-9451dfe8561e","$":"https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/e3f0f740-6fc8-415b-bba3-6a248bc21de2/track.jpg"}}},{"index":89,"time":5099640,"duration":9720,"relevance":-1,"hit":false,"text":"TU The optimal policies can be modelled as belief MDP QHCI cm I u l v H i Poupart P VIaSSIS N Hoey J Regan K An analytic solution to discrete BayeSIan reinforcement learning Proceedings of the 23rd international conference on Machine learning 2006 ACM 697 704 Nolzir glv Grove 51 m9 m9 at","previews":{"preview":{"ref":"45210525-a5a8-4bab-a0a7-9451dfe8561e","$":"https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/7020a465-0d40-4eb6-8454-6ecbcd819a40/track.jpg"}}},{"index":90,"time":5109360,"duration":20240,"relevance":-1,"hit":false,"text":"LU Applications QHCI I Clinical trials potential treatments for a disease to select from new patients or patient category at each round see c zcc oath n 3 3 zic annea r a t e e t a 5 is e 3 a at e at I Games Different moves at each round e g GO I Adaptive routing finding alternative paths also finding alternative roads for driving from A to B I Advertisement placemonts selection of an ad to display at the Webpage out of a finite set which can vary over time for each new Web page visitor Nolzr gthvouv 5 m9 049 as","previews":{"preview":{"ref":"45210525-a5a8-4bab-a0a7-9451dfe8561e","$":"https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/2f98a0d2-63ce-4bd0-bed8-56286b2bc93c/track.jpg"}}},{"index":91,"time":5129600,"duration":22960,"relevance":-1,"hit":false,"text":"TU Multi Armed Bandits and Clinical Trials cm hnp forIuneicom ZO1S 10 26 cancer clInicaI trial belmont repon Nolzr glvaouD s 709 04905","previews":{"preview":{"ref":"45210525-a5a8-4bab-a0a7-9451dfe8561e","$":"https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/b812d83e-361b-4812-82cb-e092240809af/track.jpg"}}},{"index":92,"time":5152560,"duration":41360,"relevance":-1,"hit":false,"text":"TU Example Clinical Trials in rare diseases QHCI cm Limitations of drug design for rare diseases due to I Lack of understanding of the underlying principles of the rare disease I Motivation Research advances I Unbalanced economic motivation cost benefit I Motivation Orphan Drug Act and other regulations I Unavailability of patients for standardtrials I This is the true bottleneck Villar S 5 Bowden J Wason J 2015 Mum armed Bandit Models for the Optimal Desngn of Clinical Trials Benefits and Challenges 199 215 don 10 1214 14 ST5504 Noizr glvaouD 54 m9 549 as","previews":{"preview":{"ref":"45210525-a5a8-4bab-a0a7-9451dfe8561e","$":"https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/bf4d1411-16c2-48b3-b9eb-4e720366858f/track.jpg"}}},{"index":93,"time":5193920,"duration":56640,"relevance":-1,"hit":false,"text":"TU Why should Bandit Strategies be of help here QHCI cm I The goal of Standard Randomized Controlled Trials RCT are a controlled learning setting I Control for Type and Type II errors dependent of trial size pm I In the case if the patient population N is smaller than the trial size HR T underpowered trial problem I If we change the goal to I learning sufficient to treat N as effectively as possible I then bandit strategies optimal policy for max the expected reward are perfectly suited 0 Kuleshov V Precup D 2014 Algorithms for mum armed bandit problems arXiv 1402 6028 Holznglmauv as m9 m9 at","previews":{"preview":{"ref":"45210525-a5a8-4bab-a0a7-9451dfe8561e","$":"https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/e3960148-e0b0-45c3-84ac-b8fc8bcd4ea1/track.jpg"}}},{"index":94,"time":5250560,"duration":8400,"relevance":-1,"hit":false,"text":"TU Learning vs Earning dilemma QHCI cm I Learning experimenting with all treatments I Earning selecting one treatment only based on experimentation results I Question 1 How much learning is best for an optimal treatment of N patients I Suppose N patients with a rare disease I Experimental Group E and control group C I e g control response rate pc and little information about experimental group I Question 2 How many allocations of treatment to E are necessary how much experimentation 709 L149 as News amp is","previews":{"preview":{"ref":"45210525-a5a8-4bab-a0a7-9451dfe8561e","$":"https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/7a161758-64bf-4e3e-97f9-825009822213/track.jpg"}}},{"index":95,"time":5258960,"duration":7760,"relevance":-1,"hit":false,"text":"TU Some Statistics on rare diseases orphan diseases QHCI cm I 7 000 different types more being discovered every day I 10 of the world population is suffering if all of the people with rare diseases lived in one country it would be the world s 3rd most populous country I 80 of rare diseases are genetic so are present throughout a person s lifetime even if symptoms do not immediately appear I 50 of the people affected by rare diseases are children I Are responsible for 35 of deaths in the first year of life I The prevalence distribution is skewed 80 of all rare disease patients are affected by 350 rare diseases I 50 of rare diseases do not hale a disease specific foundation supporting or researching their rare disease https g obaIgenesorg rare dIseases facts statistics https www honrch HONselect RareDIseases Nelmgu Gveuo as 709 049 as","previews":{"preview":{"ref":"45210525-a5a8-4bab-a0a7-9451dfe8561e","$":"https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/8451a69f-11ad-4902-a25a-d412b52cf810/track.jpg"}}},{"index":96,"time":5266720,"duration":20480,"relevance":-1,"hit":false,"text":"IU Example Rare Disease CADASIL 9H5 Cerebral Autosomal Dominant Arteriopathy with Subcortical Infarcts and Leukoencephalopathy is a hereditary disease affecting all the small cerebral arteries It causes subcortical infarcts and damages the white matter leukoencephalopathy and it is due to various mutations of the Notch3 gene Situated on chromosome 19 Joutel A et al 1996 thchB mutations ll l CADASIL a hereditary adult onset coriditioniausmg stroke and dementia Nature 383 6602 707 710 dOI 10 1038 383707a0 Nclzmglv Gvoup 59 m9 549 as","previews":{"preview":{"ref":"45210525-a5a8-4bab-a0a7-9451dfe8561e","$":"https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/bbb46923-b591-4330-ab8f-17a2e2d717c1/track.jpg"}}},{"index":97,"time":5287200,"duration":24360,"relevance":-1,"hit":false,"text":"TU CADASIL QHCI cm ChabrIaK H Joute A HDIchgans M Tournier Lasserve E Bousser M G 2009 CADASIL The Lancet Neurology 8 7 643 653 d0I hnp dx d0I org 10 1016 51474 4422 09 70127 9 ch new we 0 qu","previews":{"preview":{"ref":"45210525-a5a8-4bab-a0a7-9451dfe8561e","$":"https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/a1838e06-c47e-42bc-a268-38734a107ef3/track.jpg"}}},{"index":98,"time":5311560,"duration":5800,"relevance":-1,"hit":false,"text":"QHCI Conclusion and Future Challenges","previews":{"preview":{"ref":"45210525-a5a8-4bab-a0a7-9451dfe8561e","$":"https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/a521675a-8f35-40a8-acd3-8fe6991f0a3f/track.jpg"}}},{"index":99,"time":5317360,"duration":81280,"relevance":-1,"hit":false,"text":"TU Final conclusion QHCI cm Bandit strategy ls experimentation worth it for a small number N I Reconcile clinical trials and clinical practice Extensions should deal with randomization delayed responses and uncertainty around N Bayesian bandits need Online ML Bandits are a great source of inspirations and building blocks for solving many problems Future work convex optimization contextual combinatorial 0 Berry D A Fristedt BV 1985 Bandit problems sequential allocation of experiments Monographs on statistics and applied probability Springer News Group 93 m9 549 as","previews":{"preview":{"ref":"45210525-a5a8-4bab-a0a7-9451dfe8561e","$":"https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/71f34059-5140-44d2-bfc5-81624434cbd0/track.jpg"}}},{"index":100,"time":5398640,"duration":16347,"relevance":-1,"hit":false,"text":"V 7 m 9H ADA I 1 x v f I s Thankyou Nmmgu Evans 94 ms 549 as","previews":{"preview":{"ref":"45210525-a5a8-4bab-a0a7-9451dfe8561e","$":"https://tube.tugraz.at/static/mh_default_org/engage-player/e85fac19-c0af-4517-88ed-f961797258d8/8e98bc46-c08d-4e0a-a37b-f8b7e96199f1/track.jpg"}}}]}}}}</article></body></html>