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:20231215T160500
DTEND;TZID=Australia/Melbourne:20231215T161500
UID:siggraphasia_SIGGRAPH Asia 2023_sess159_papers_397@linklings.com
SUMMARY:Compact Neural Graphic Primitives with Learned Hash Probing
DESCRIPTION:Technical Papers\n\nTowaki Takikawa (NVIDIA, University of Tor
 onto); Thomas Müller, Merlin Nimier-David, and Alex Evans (NVIDIA); Sanja 
 Fidler (NVIDIA, University of Toronto); Alec Jacobson (University of Toron
 to, Adobe Research); and Alexander Keller (NVIDIA)\n\nNeural graphics prim
 itives are faster and achieve higher quality when their neural networks ar
 e augmented by spatial data structures that hold trainable features arrang
 ed in a grid. However, existing feature grids either come with a large mem
 ory footprint (dense or factorized grids, trees, and hash tables) or slow 
 performance (index learning and vector quantization). In this paper, we sh
 ow that a hash table with learned probes has neither disadvantage, resulti
 ng in a favorable combination of size and speed. Inference is faster than 
 unprobed hash tables at equal quality while training is only 1.2-2.6x slow
 er, significantly outperforming prior index learning approaches. We arrive
  at this formulation by casting all feature grids into a common framework:
  they each correspond to a lookup function that indexes into a table of fe
 ature vectors. In this framework, the lookup functions of existing data st
 ructures can be combined by simple arithmetic combinations of their indice
 s, resulting in Pareto optimal compression and speed.\n\nRegistration Cate
 gory: Full Access\n\nSession Chair: Fei Hou (Institute of Software, Chines
 e Academy of Sciences; University of Chinese Academy of Sciences)
URL:https://asia.siggraph.org/2023/full-program?id=papers_397&sess=sess159
END:VEVENT
END:VCALENDAR