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:20260422T000711Z
LOCATION:501-502
DTSTART;TZID=America/Denver:20231113T140000
DTEND;TZID=America/Denver:20231113T142000
UID:submissions.supercomputing.org_SC23_sess440_ws_ai4s112@linklings.com
SUMMARY:Accelerating Particle and Fluid Simulations with Differentiable an
 d Interpretable Graph Networks for Solving Forward and Inverse Problems
DESCRIPTION:Krishna Kumar (University of Texas System) and Yonjin Choi (Un
 iversity of Texas)\n\nWe leverage physics-embedded differentiable graph ne
 twork simulators (GNS) to accelerate particulate and fluid simulations to 
 solve forward and inverse problems. GNS represents the domain as a graph w
 ith particles as nodes and learned interactions as edges, improving genera
 lization to new environments. GNS achieves over 165x speedup for granular 
 flow prediction compared to parallel CPU simulations. We propose a novel h
 ybrid GNS/Material Point Method to accelerate forward simulations by minim
 izing error on a surrogate model, achieving 24x speedup. The differentiabl
 e GNS enables solving inverse problems through automatic differentiation, 
 identifying material parameters that result in target runout distances. We
  demonstrate solving inverse problems by iteratively updating the friction
  angle by computing the gradient of a loss function based on the final and
  target runouts, thereby identifying the friction angle that matches the o
 bserved runout. The physics-embedded and differentiable simulators open an
  exciting paradigm for AI-accelerated design, control, and optimization.\n
 \nTag: Artificial Intelligence/Machine Learning\n\nRegistration Category: 
 Workshop Reg Pass\n\nSession Chairs: Murali Emani (Argonne National Labora
 tory (ANL)); Gokcen Kestor (Barcelona Supercomputing Center (BSC); Univers
 ity of California, Merced); and Dong Li (University of California, Merced)
 \n\n
END:VEVENT
END:VCALENDAR