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:20210916T132530Z
LOCATION:Ella Maillart
DTSTART;TZID=Europe/Stockholm:20210706T163000
DTEND;TZID=Europe/Stockholm:20210706T173000
UID:submissions.pasc-conference.org_PASC21_sess171_panel101@linklings.com
SUMMARY:Learning Versus Understanding: Is AI a Definitive Game Changer in 
 Science?
DESCRIPTION:Panel\n\nLearning Versus Understanding: Is AI a Definitive Gam
 e Changer in Science?\n\nGirone, Estrada, Fleuret, Konstantinidis, Sanyal\
 n\nArtificial intelligence and machine learning have rapidly become a must
  across the breadth of scientific research activities, as well as opening 
 new directions in emerging computational disciplines. Can this be attribut
 ed to a failure of the traditional scientific methodology or is it due to 
 a synergy of rapid HPC development with a data deluge that cannot be explo
 ited otherwise?<br /><br />AI and ML can automate the scientific method (h
 ypothesis generation, experiment design, data collection, analysis and inf
 erence). In particular there has been a lot of recent activity in hypothes
 is generation with generative models/causal learning/knowledge representat
 ion. AI may "augment" humans through efficient collaboration to achieve tr
 ue collective intelligence, overcoming human flaws (narrow framing of a pr
 oblem, personal bias, ego, pride and investment in "sunken cost," and so o
 n). How can we identify the areas of research with greatest opportunities 
 to exploit this potential and ensure an optimum combination of talent and 
 tools to yield an efficient accumulation of wisdom?<br /><br />Can ML/DL f
 ully replace a knowledge-based analytical or numerical solution of a physi
 cal problem through a "black-box"? Is that more efficient and how can trus
 t be maintained?<br /><br />The training of AI systems has very high costs
  in terms of computing resources, energy cost and related CO2 production. 
 Do the benefits outweigh these costs? Is it sustainable and what might a g
 reen AI/ML look like?<br /><br />Can scientists combine traditional scient
 ific and computational methodologies with AI to achieve the best of both w
 orlds?
END:VEVENT
END:VCALENDAR