BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:Energy-conserving equivariant GNN predictions of stiffness for lat
 tice materials   - Ivan Grega\, CUED
DTSTART:20240126T160000Z
DTEND:20240126T170000Z
UID:TALK209665@talks.cam.ac.uk
CONTACT:46601
DESCRIPTION:Lattices emerged in recent decades as a promising class of arc
 hitected materials with a vast design space. Many machine learning models 
 have been proposed as surrogate to numerical modelling in predicting their
  mechanical properties for rapid design applications. However\, they are o
 ften not scalable\, lack the appropriate physical constraints and hence ar
 e limited to a small fragment of the vast design space. Here we develop a 
 graph based neural network to predict the fourth-order stiffness tensor of
  any arbitrary periodic lattice. We build upon the equivariant MACE model 
 (Batatia\, Kovács\, Csányi et al.) and introduce positive semi-definite 
 constraints that ensure energy conservation. We trained the model on a gen
 eralised dataset of unit cells and demonstrate an example application of t
 he model in structural optimization.
LOCATION:JDB Seminar Room\, CUED
END:VEVENT
END:VCALENDAR