BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Memento EPFL//
BEGIN:VEVENT
SUMMARY:Machine Learning Materials Modelling
DTSTART:20260427T150000
DTEND:20260427T170000
DTSTAMP:20260427T221925Z
UID:0ee66157ccd195dbb9a010dcafc58043824d0b645aa7a29828065ca7
CATEGORIES:Conferences - Seminars
DESCRIPTION:Dr. Jonathan Schmid \nLarge-scale machine learning (ML) model
 s are transforming materials modelling by dramatically increasing the scal
 e and speed at which materials can be explored. A central example of this 
 development is the Alexandria database\, which we created through high-thr
 oughput density functional theory (DFT) searches accelerated by crystal-gr
 aph-attention networks. With millions of DFT-relaxed crystal structures sp
 anning one-\, two-\, and three-dimensional materials\, Alexandria multipli
 es the number of known stable materials and provides a foundation for trai
 ning next-generation universal ML force fields.\nThe talk will further exp
 lore how ML methods help bridge the gap between ab initio simulations and 
 real experimental conditions. Examples include interpretable ML models tha
 t correct systematic errors in density functional theory and ML force-fiel
 d–accelerated simulations that enable the study of materials questions p
 reviously beyond computational reach. Together\, these advances highlight 
 the transformative impact of ML on modern materials science.
LOCATION:PH H3 31 https://plan.epfl.ch/?room==PH%20H3%2031
STATUS:CONFIRMED
END:VEVENT
END:VCALENDAR