BEGIN:VCALENDAR
VERSION:2.0
PRODID:Linklings LLC
BEGIN:VTIMEZONE
TZID:Australia/Melbourne
X-LIC-LOCATION:Australia/Melbourne
BEGIN:DAYLIGHT
TZOFFSETFROM:+1000
TZOFFSETTO:+1100
TZNAME:AEDT
DTSTART:19721003T020000
RRULE:FREQ=YEARLY;BYMONTH=4;BYDAY=1SU
END:DAYLIGHT
BEGIN:STANDARD
DTSTART:19721003T020000
TZOFFSETFROM:+1100
TZOFFSETTO:+1000
TZNAME:AEST
RRULE:FREQ=YEARLY;BYMONTH=10;BYDAY=1SU
END:STANDARD
END:VTIMEZONE
BEGIN:VEVENT
DTSTAMP:20240214T070241Z
LOCATION:Darling Harbour Theatre\, Level 2 (Convention Centre)
DTSTART;TZID=Australia/Melbourne:20231212T093000
DTEND;TZID=Australia/Melbourne:20231212T124500
UID:siggraphasia_SIGGRAPH Asia 2023_sess209_papers_238@linklings.com
SUMMARY:Neural Metamaterial Networks for Nonlinear Material Design
DESCRIPTION:Technical Papers\n\nYue Li, Stelian Coros, and Bernhard Thomas
 zewski (ETH Zürich)\n\nNonlinear metamaterials with tailored mechanical pr
 operties have applications in engineering, medicine, robotics, and beyond.
  While modeling\ntheir macromechanical behavior is challenging in itself, 
 finding structure\nparameters that lead to an ideal approximation of high-
 level performance goals\nis a daunting task. In this work, we propose Neur
 al Metamaterial Networks\n(NMN)—smooth neural representations that encode 
 the nonlinear mechanics of entire metamaterial families. Given structure p
 arameters as input,\nNMN return continuously differentiable strain energy 
 density functions,\nthus guaranteeing conservative forces by construction.
  Though trained on\nsimulation data, NMN do not inherit the discontinuitie
 s resulting from topological changes in finite element meshes. They instea
 d provide a smooth\nmap from parameter to performance space that is fully 
 differentiable and\nthus well-suited for gradient-based optimization. On t
 his basis, we formulate\ninverse material design as a nonlinear programmin
 g problem that leverages\nneural networks for both objective functions and
  constraints. We use this\napproach to automatically design materials with
  desired strain-stress curves,\nprescribed directional stiffness, and Pois
 son ratio profiles. We furthermore\nconduct ablation studies on network no
 nlinearities and show the advantages\nof our approach compared to native-s
 cale optimization.\n\nRegistration Category: Full Access, Enhanced Access,
  Trade Exhibitor, Experience Hall Exhibitor
URL:https://asia.siggraph.org/2023/full-program?id=papers_238&sess=sess209
END:VEVENT
END:VCALENDAR