BEGIN:VCALENDAR
VERSION:2.0
PRODID:Linklings LLC
BEGIN:VTIMEZONE
TZID:America/Denver
X-LIC-LOCATION:America/Denver
BEGIN:DAYLIGHT
TZOFFSETFROM:-0700
TZOFFSETTO:-0600
TZNAME:MDT
DTSTART:19700308T020000
RRULE:FREQ=YEARLY;BYMONTH=3;BYDAY=2SU
END:DAYLIGHT
BEGIN:STANDARD
TZOFFSETFROM:-0600
TZOFFSETTO:-0700
TZNAME:MST
DTSTART:19701101T020000
RRULE:FREQ=YEARLY;BYMONTH=11;BYDAY=1SU
END:STANDARD
END:VTIMEZONE
BEGIN:VEVENT
DTSTAMP:20260422T000713Z
LOCATION:501-502
DTSTART;TZID=America/Denver:20231114T110000
DTEND;TZID=America/Denver:20231114T113000
UID:submissions.supercomputing.org_SC23_sess296_gb108@linklings.com
SUMMARY:Scaling the Leading Accuracy of Deep Equivariant Models to Biomole
 cular Simulations of Realistic Size
DESCRIPTION:Boris Kozinsky, Albert Musaelian, Anders Johansson, and Simon 
 Batzner (Harvard University)\n\nThis work brings the leading accuracy, sam
 ple efficiency, and robustness of deep equivariant neural networks to the 
 extreme computational scale. This is achieved through a combination of inn
 ovative model architecture, massive parallelization, and models and implem
 entations optimized for efficient GPU utilization.  The resulting Allegro 
 architecture bridges the accuracy/speed tradeoff of atomistic simulations 
 and enables description of dynamics in structures of unprecedented complex
 ity at quantum fidelity. To illustrate the scalability of Allegro, we perf
 orm nanoseconds-long stable simulations of protein dynamics and scale up t
 o a 44-million atom structure of a complete, all-atom, explicitly solvated
  HIV capsid on the Perlmutter supercomputer.  We demonstrate excellent str
 ong scaling up to 100 million atoms and 70% weak scaling to 5120 A100 GPUs
 \n\nRegistration Category: Tech Program Reg Pass\n\nSession Chair: Taisuke
  Boku (University of Tsukuba, Advanced HPC‑AI Research and Development Sup
 port Center (HAIRDESC))\n\n
END:VEVENT
END:VCALENDAR