BEGIN:VCALENDAR
VERSION:2.0
PRODID:Linklings LLC
BEGIN:VTIMEZONE
TZID:Asia/Tokyo
X-LIC-LOCATION:Asia/Tokyo
BEGIN:STANDARD
TZOFFSETFROM:+0900
TZOFFSETTO:+0900
TZNAME:JST
DTSTART:18871231T000000
END:STANDARD
END:VTIMEZONE
BEGIN:VEVENT
DTSTAMP:20250110T023313Z
LOCATION:Hall B7 (1)\, B Block\, Level 7
DTSTART;TZID=Asia/Tokyo:20241206T092300
DTEND;TZID=Asia/Tokyo:20241206T093400
UID:siggraphasia_SIGGRAPH Asia 2024_sess141_papers_734@linklings.com
SUMMARY:Trust-Region Eigenvalue Filtering for Projected Newton
DESCRIPTION:Technical Papers\n\nHonglin Chen (Columbia University); Hsueh-
 Ti Derek Liu (Roblox, University of British Columbia); Alec Jacobson (Univ
 ersity of Toronto, Adobe Research); David I.W. Levin (University of Toront
 o, NVIDIA); and Changxi Zheng (Columbia University)\n\nWe introduce a nove
 l adaptive eigenvalue filtering strategy to stabilize and accelerate the o
 ptimization of Neo-Hookean energy and its variants under the Projected New
 ton framework. For the first time, we show that Newton’s method, Projected
  Newton with eigenvalue clamping and Projected Newton with absolute eigenv
 alue filtering can be unified using ideas from the\ngeneralized trust regi
 on method. Based on the trust-region fit, our model adaptively chooses the
  correct eigenvalue filtering strategy to apply during the optimization. O
 ur method is simple but effective, requiring only two lines of code change
  in the existing Projected Newton framework. We validate our model outperf
 orms stand-alone variants across a number of experiments\non quasistatic s
 imulation of deformable solids over a large dataset.\n\nRegistration Categ
 ory: Full Access, Full Access Supporter\n\nLanguage Format: English Langua
 ge\n\nSession Chair: Sheldon Andrews (École de technologie supérieure (ÉTS
 ))
URL:https://asia.siggraph.org/2024/program/?id=papers_734&sess=sess141
END:VEVENT
END:VCALENDAR