BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//pretalx//cfp.scipy.org//2023//speaker//TEKNGX
BEGIN:VTIMEZONE
TZID:America/Chicago
BEGIN:DAYLIGHT
DTSTART:20220714T000000
TZNAME:CDT
TZOFFSETFROM:-0500
TZOFFSETTO:-0500
END:DAYLIGHT
BEGIN:STANDARD
DTSTART:20221106T020000
RDATE:20231105T020000
TZNAME:CST
TZOFFSETFROM:-0500
TZOFFSETTO:-0600
END:STANDARD
BEGIN:DAYLIGHT
DTSTART:20230312T030000
RDATE:20240310T030000
TZNAME:CDT
TZOFFSETFROM:-0600
TZOFFSETTO:-0500
END:DAYLIGHT
END:VTIMEZONE
BEGIN:VEVENT
SUMMARY:Allegro and FLARE: Fast and accurate machine learning potentials f
 or extreme-scale simulations - Anders  Johansson
DTSTART;TZID=America/Chicago:20230714T131500
DTEND;TZID=America/Chicago:20230714T134500
DTSTAMP:20261007T035725Z
UID:pretalx-2023-BDV3EE@cfp.scipy.org
DESCRIPTION:Allegro and FLARE are two very different packages for construc
 ting machine learning potentials that are fast\, accurate\, and suitable f
 or extreme-scale molecular dynamics simulations. Allegro uses PyTorch for 
 efficient equivariant potentials with state-of-the-art accuracy\, while FL
 ARE is a sparse Gaussian process potential with an optimized C++ training 
 backend leveraging Kokkos\, OpenMP\, and MPI for state-of-the-art performa
 nce\, and a user-friendly Python frontend. We will compare and contrast th
 e two methods\, discuss lessons learned\, and show spectacular scientific 
 applications.
LOCATION:Grand Salon C
URL:https://cfp.scipy.org/2023/talk/BDV3EE/
END:VEVENT
END:VCALENDAR