BEGIN:VCALENDAR
VERSION:2.0
PRODID:Linklings LLC
BEGIN:VTIMEZONE
TZID:Asia/Hong_Kong
X-LIC-LOCATION:Asia/Hong_Kong
BEGIN:STANDARD
TZOFFSETFROM:+0800
TZOFFSETTO:+0800
TZNAME:HKT
DTSTART:19911015T033000
END:STANDARD
END:VTIMEZONE
BEGIN:VEVENT
DTSTAMP:20251218T030656Z
LOCATION:Meeting Room S426+S427\, Level 4
DTSTART;TZID=Asia/Hong_Kong:20251217T133100
DTEND;TZID=Asia/Hong_Kong:20251217T134200
UID:siggraphasia_SIGGRAPH Asia 2025_sess139_papers_1659@linklings.com
SUMMARY:Harmonic Caching for Walk on Spheres
DESCRIPTION:Zihong Zhou (Dartmouth College), Eugene d'Eon and Rohan Sawhne
 y (NVIDIA), and Wojciech Jarosz (Dartmouth College)\n\nWe present a varian
 ce reduction technique for Walk on Spheres (WoS) that solves elliptic part
 ial differential equations (PDEs) by combining overlapping harmonic expans
 ions of the solution, each estimated using unbiased Monte Carlo walks. Our
  method supports both the Laplace and screened-Poisson equations with Diri
 chlet, Neumann and Robin boundary conditions in 2D and 3D. By adaptively c
 overing the domain with local expansion regions and extrapolating from eac
 h using a truncated Fourier basis, we achieve orders of magnitude lower er
 ror than traditional pointwise WoS, in equal time. While low-order truncat
 ions could achieve low bias, unbiased reconstruction is possible with stoc
 hastic truncation. Compared to other recently developed caching algorithms
  for WoS such as Boundary and Mean Value Caching, our approach generally g
 enerates results with lower error and fewer correlation artifacts.\n\nRegi
 stration Category: Full Access, Full Access Supporter\n\nSession Chair: Sh
 eng Li (Peking University)\n\n
END:VEVENT
END:VCALENDAR