BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:Rapid Discovery of Novel Materials by Coordinate-free Coarse Grain
 ing using Wyckoff Representations - Rhys Goodall 
DTSTART:20210726T153000Z
DTEND:20210726T160000Z
UID:TALK161449@talks.cam.ac.uk
CONTACT:Bingqing Cheng
DESCRIPTION:Coarse-graining to molecular graphs is critical to ML-driven m
 olecular discovery. We show how Wyckoff representations - coordinate-free 
 sets of symmetry-related positions in crystals - can empower analogous imp
 rovements in inorganic material discovery. Unlike atomic positions\, Wycko
 ff representations are computably enumerable enabling screening campaigns 
 across novel materials space. Critically\, compared to composition-based a
 pproaches we avoid the majority of the cost of structure searching when va
 lidating predictions. We propose "Wren" - a ML model operating on Wyckoff 
 representations. We demonstrate Wren's potential by identifying 1\,558 nov
 el materials that lie below the convex hull of previously calculated mater
 ials from just 5\,675 ab-initio calculations.\narXiv pre-print: https://ar
 xiv.org/abs/2106.11132
LOCATION:virtual ZOOM meeting ID: 263 591 6003\, Passcode: 000042\, https:
 //us02web.zoom.us/j/2635916003?pwd=ZlBEQnRENGwxNmJGMENGMWxjak5nUT09
END:VEVENT
END:VCALENDAR