BEGIN:VCALENDAR
VERSION:2.0
PRODID:Linklings LLC
BEGIN:VTIMEZONE
TZID:Europe/Stockholm
X-LIC-LOCATION:Europe/Stockholm
BEGIN:DAYLIGHT
TZOFFSETFROM:+0100
TZOFFSETTO:+0200
TZNAME:CEST
DTSTART:19700308T020000
RRULE:FREQ=YEARLY;BYMONTH=3;BYDAY=-1SU
END:DAYLIGHT
BEGIN:STANDARD
TZOFFSETFROM:+0200
TZOFFSETTO:+0100
TZNAME:CET
DTSTART:19701101T020000
RRULE:FREQ=YEARLY;BYMONTH=10;BYDAY=-1SU
END:STANDARD
END:VTIMEZONE
BEGIN:VEVENT
DTSTAMP:20241120T082408Z
LOCATION:HG F 26.5
DTSTART;TZID=Europe/Stockholm:20240603T113000
DTEND;TZID=Europe/Stockholm:20240603T120000
UID:submissions.pasc-conference.org_PASC24_sess142_msa200@linklings.com
SUMMARY:Infrastructure to Support a Community of Drug Response Prediction 
 Modelers
DESCRIPTION:Minisymposium\n\nJustin Wozniak (Argonne National Laboratory, 
 University of Chicago)\n\nThe intersection of precision medicine and machi
 ne learning (ML) offers a wide range of problems and possible approaches. 
  The prediction of tumor response to single and combination drug agents is
  an active area of ML application development, as tens of deep learning mo
 dels are currently available from the community and are under active devel
 opment.  Comparing the behavior of these models is very difficult, and is 
 not a well-studied area, as different projects differ wildly in their prob
 lem assumptions and approach to the problem.  A range of other problems mu
 st also be addressed, including assessing the robustness of the models acr
 oss a range of health science use cases, hardware resources, and other sit
 uations. In this presentation, we will describe our approach to build infr
 astructure to support the studies outlined above.  We are developing a sca
 lable workflow framework to manage, curate, and execute community models i
 n varying scientific problems.  Typical use cases include hyperparameter o
 ptimization to tune models for general or specific use cases, comparison t
 o study the differences across models, and cross-study analyses that compa
 re training datasets.  This presentation will cover the internal design of
  the system, how it may be extended and used, and results from supercomput
 ers Polaris and Aurora.\n\nDomain: Life Sciences, Computational Methods an
 d Applied Mathematics\n\nSession Chair: Thomas Brettin (Argonne National L
 aboratory, University of Chicago)
END:VEVENT
END:VCALENDAR