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:20210916T132453Z
LOCATION:Louis Favre
DTSTART;TZID=Europe/Stockholm:20210708T170000
DTEND;TZID=Europe/Stockholm:20210708T173000
UID:submissions.pasc-conference.org_PASC21_sess197_msa259@linklings.com
SUMMARY:On the Role of AI in the High Throughput Application’s Life Cycle
DESCRIPTION:Minisymposium\n\nOn the Role of AI in the High Throughput Appl
 ication’s Life Cycle\n\nEstrada\n\nOver the past years, the use of AI has 
 become ubiquitous in most disciplines, and High Throughput applications ar
 e not the exception. AI increasingly plays a central role across the whole
  workflow pipeline, from workload forecasting, adaptive scheduling, self-m
 anaged resource allocation, and on the fly analysis. As we consider a path
 way towards reproducible, scalable, and trustworthy science, we must pay s
 pecial attention to the impact of AI in HTC, and how current practices can
  advance or hinder these efforts. In this talk we present three case 
 studies, where AI and HTC coexist and highlight one or more of the themes 
 of our roadmap. 1) Scalability: we show how data representation enables sc
 alable high throughput in-situ analysis of protein trajectories. 2) Reprod
 ucibility: we outline a study of workload variability and its impact on pe
 rformance forecasting at scale. 3) Trust: we present pitfalls of model sel
 ection in medical image analysis.\n\nDomain: CS and Math, Emerging Applica
 tions, Chemistry and Materials, Climate and Weather, Physics
END:VEVENT
END:VCALENDAR