BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:Graph Neural Networks for Biomedical Data - Marinka Zitnik\, Harva
 rd University
DTSTART:20201125T150000Z
DTEND:20201125T160000Z
UID:TALK152932@talks.cam.ac.uk
CONTACT:Mateja Jamnik
DESCRIPTION:"Join us on Zoom webinar":https://cl-cam-ac-uk.zoom.us/j/91253
 900399?pwd=SU5TNnpYdDlQbzQ4SEVPVWVWa0Nldz09\n\nThe success of machine lear
 ning depends heavily on the choice of representations used for downstream 
 tasks. Graph neural networks have emerged as a predominant choice for lear
 ning representations of networked data. Still\, methods require abundant l
 abel information and focus either on nodes or entire graphs. In this talk\
 , I describe our efforts to expand the scope and ease the applicability of
  graph representation learning. First\, I outline SubGNN\, the first subgr
 aph neural network for learning disentangled subgraph representations. Sec
 ond\, I will describe G-Meta\, a novel meta-learning approach for graphs. 
 G-Meta uses subgraphs to generalize to completely new graphs and never-bef
 ore-seen labels using only a handful of nodes or edges. G-Meta is theoreti
 cally justified and scales to orders of magnitude larger datasets than pri
 or work. Finally\, I will discuss applications in biology and medicine. Th
 e new methods have enabled the repurposing of drugs for new diseases\, inc
 luding COVID-19\, where our predictions were experimentally verified in th
 e wet laboratory. Further\, the methods enabled discovering dozens of comb
 inations of drugs safe for patients with considerably fewer unwanted side 
 effects than today's treatments. The methods also allow for molecular phen
 otyping\, much better than more complex algorithms. Lastly\, I describe ou
 r efforts in learning actionable representations that allow users of our m
 odels to receive predictions that can be interpreted meaningfully. \n\n*BI
 O:*\nMarinka Zitnik is an Assistant Professor at Harvard University with a
 ppointments in the Department of Biomedical Informatics\, Blavatnik Instit
 ute\, Broad Institute of MIT and Harvard\, and Harvard Data Science. Dr. Z
 itnik is a computer scientist studying machine learning\, focusing on chal
 lenges brought forward by data in science\, medicine\, and health. She has
  published extensively on representation learning\, knowledge graphs\, dat
 a fusion\, graph ML (NeurIPS\, JMLR\, IEEE TPAMI\, KDD\, ICLR)\, and appli
 cations to biomedicine (Nature Methods\, Nature Communications\, PNAS). He
 r algorithms are used by major institutions\, including Baylor College of 
 Medicine\, Karolinska Institute\, Stanford Medical School\, and Massachuse
 tts General Hospital. Her work received several best paper\, poster\, and 
 research awards from the International Society for Computational Biology. 
 She has recently been named a Rising Star in Electrical Engineering and Co
 mputer Science (EECS) by MIT and also a Next Generation in Biomedicine by 
 the Broad Institute\, being the only young scientist who received such rec
 ognition in both EECS and Biomedicine.\n\n
LOCATION:Zoom
END:VEVENT
END:VCALENDAR