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:20231113T113000
DTEND;TZID=America/Denver:20231113T115000
UID:submissions.supercomputing.org_SC23_sess440_ws_ai4s101@linklings.com
SUMMARY:A Comparison of Mesh-Free Differentiable Programming and Data-Driv
 en Strategies for Optimal Control under PDE Constraints
DESCRIPTION:Roussel Desmond Nzoyem Ngueguin, David A.W. Barton, and Tom De
 akin (University of Bristol)\n\nThe field of Optimal Control under Partial
  Differential Equations (PDE) constraints is rapidly changing under the in
 fluence of Deep Learning and the accompanying automatic differentiation li
 braries. Novel techniques like Physics-Informed Neural Networks (PINNs) an
 d Differentiable Programming (DP) are to be contrasted with established nu
 merical schemes like Direct-Adjoint Looping (DAL). We present a comprehens
 ive comparison of DAL, PINN, and DP using a general-purpose mesh-free diff
 erentiable PDE solver based on Radial Basis Functions. Under Laplace and N
 avier-Stokes equations, we found DP to be extremely effective as it produc
 es the most accurate gradients; thriving even when DAL fails and PINNs str
 uggle. Additionally, we provide a detailed benchmark highlighting the limi
 ted conditions under which any of those methods can be efficiently used. O
 ur work provides a guide to Optimal Control practitioners and connects the
 m further to the Deep Learning community.\n\nTag: Artificial Intelligence/
 Machine Learning\n\nRegistration Category: Workshop Reg Pass\n\nSession Ch
 airs: Murali Emani (Argonne National Laboratory (ANL)); Gokcen Kestor (Bar
 celona Supercomputing Center (BSC); University of California, Merced); and
  Dong Li (University of California, Merced)\n\n
END:VEVENT
END:VCALENDAR