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:20240214T070250Z
LOCATION:Meeting Room C4.8\, Level 4 (Convention Centre)
DTSTART;TZID=Australia/Melbourne:20231215T143000
DTEND;TZID=Australia/Melbourne:20231215T144000
UID:siggraphasia_SIGGRAPH Asia 2023_sess171_papers_333@linklings.com
SUMMARY:LiCROM: Linear-Subspace Continuous Reduced Order Modeling with Neu
 ral Fields
DESCRIPTION:Technical Papers\n\nYue Chang (University of Toronto), Peter Y
 ichen Chen (MIT CSAIL), Zhecheng Wang (University of Toronto), Maurizio M.
  Chiaramonte and Kevin Carlberg (Meta Reality Labs Research), and Eitan Gr
 inspun (University of Toronto)\n\nLinear reduced-order modeling (ROM) simp
 lifies complex simulations by approximating the behavior of a system using
  a simplified kinematic representation. Typically, ROM\nis trained on inpu
 t simulations created with a specific spatial discretization, \nand then s
 erves to accelerate simulations with the same discretization. \nThis discr
 etization-dependence is restrictive. \n\nBecoming independent of a specifi
 c discretization would provide flexibility to mix and match  mesh resoluti
 ons, connectivity, and type (tetrahedral, hexahedral) in training data; to
  \naccelerate simulations with novel discretizations unseen during trainin
 g; \nand to accelerate adaptive simulations that temporally or parametrica
 lly change\nthe discretization. \n\nWe present a flexible, discretization-
 independent approach to reduced-order modeling. \nLike traditional ROM, we
  represent the configuration as a linear combination of displacement\nfiel
 ds. Unlike traditional ROM, our displacement fields are continuous maps fr
 om every point on the reference domain to a corresponding displacement vec
 tor; these maps are\nrepresented as implicit neural fields.\n\nWith linear
  continuous ROM (LiCROM), our training set can include multiple geometries
  undergoing multiple loading conditions, independent of their discretizati
 on. This opens the door to novel applications of reduced order modeling. F
 or instance, we can accelerate\nsimulations on geometries unseen during tr
 aining, and simulations that modify the geometry at runtime, for instance 
 via cutting, hole punching, and even swapping the entire mesh. Indeed, we 
 achieve one-shot generalization by training on a single geometry but testi
 ng on multiple unseen geometries.\n\nRegistration Category: Full Access\n\
 nSession Chair: Qixing Huang (University of Texas at Austin)
URL:https://asia.siggraph.org/2023/full-program?id=papers_333&sess=sess171
END:VEVENT
END:VCALENDAR