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:20220812T074334Z
LOCATION:Nairobi Room
DTSTART;TZID=Europe/Stockholm:20220628T113000
DTEND;TZID=Europe/Stockholm:20220628T120000
UID:submissions.pasc-conference.org_PASC22_sess143_msa131@linklings.com
SUMMARY:Machine Learning to Investigate Material Properties: How to Improv
 e Machine Performance
DESCRIPTION:Minisymposium\n\nMachine Learning to Investigate Material Prop
 erties: How to Improve Machine Performance\n\nGagliardi\n\nMachine learnin
 g (ML) is emerging as a new tool for many different fields which now span,
  among the others, chemistry, physics and material science [1]. The idea i
 s to use ML algorithms to identify, starting from big data analysis, subtl
 e correlations between simple elemental quantities and complex material pr
 operties and then to use this to predict them. This approach can help to s
 creen many material properties directly in-silico avoiding more computatio
 nal expensive ab-initio calculations and experimental measurements. Howeve
 r, adapting existing ML architectures to problems in material science is n
 ot straightforward. Several aspects need to be addressed to improve machin
 e performance which can be summarized into prediction accuracy and general
 ization. Improving these aspects require to go into the details of the mac
 hine and analyze the way they learn from a training dataset. This allows t
 o identify which architecture, training algorithm and dataset are relevant
  for the problem at hand. In the present talk, starting from some examples
  of ML applications to material science [2], several different strategies 
 will be discussed to improve algorithm performances. <br /><br />[1] Wei L
 i, Ryan Jacobs, Dane Morgan, Computational Materials Science 150, 454-463 
 (2018).<br />[2] J. Lederer, et al., Advanced Theory and Simulations 2 (2)
 , 1800136 (2019).\n\nDomain: Chemistry and Materials, Physics
END:VEVENT
END:VCALENDAR