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:20210916T132456Z
LOCATION:
DTSTART;TZID=Europe/Stockholm:20210706T173000
DTEND;TZID=Europe/Stockholm:20210706T190000
UID:submissions.pasc-conference.org_PASC21_sess182_post112@linklings.com
SUMMARY:P05 - Trajectory-Based Machine Learning Method for Molecular Dynam
 ics
DESCRIPTION:Poster\n\nP05 - Trajectory-Based Machine Learning Method for M
 olecular Dynamics\n\nHan\n\nA trajectory-based machine learning package (T
 rajML) for molecular dynamics (MD) was developed for instant modeling of t
 he force field and prediction of molecular configurations for MD trajector
 ies. In the code, the ML features were constructed in a way that the ML pr
 ocesses were independent for each atom allowing an easy parallelization. A
 side from that, the prediction of MD steps, which is computationally inten
 sive, was programmed with CUDA and the code was run on GPUs to achieve hig
 h performance. TrajML MD consumes similar computational resources as class
 ical MD but can simulate complex systems with a higher accuracy.
END:VEVENT
END:VCALENDAR