BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:Physical Inductive Biases for World Models - Dr Ayush Tewari - Dep
 artment of Engineering\, University of Cambridge
DTSTART:20251126T150500Z
DTEND:20251126T155500Z
UID:TALK237226@talks.cam.ac.uk
CONTACT:Ben Karniely
DESCRIPTION:Recent advances in world models for vision have been driven la
 rgely by powerful 2D generative models that predict future scene states fr
 om visual observations and planned actions\, often with little or no built
 -in knowledge of 3D geometry or physical dynamics. In this talk\, I will d
 iscuss why physical inductive biased\, such as explicit 3D structure\, rem
 ain essential for world modelling. I will present recent work showing how 
 incorporating geometric and physical priors into learned models leads to m
 ore generalisable behaviour\, significantly more efficient inference\, and
  better control over predictions.\n\nLink to join virtually: https://cam-a
 c-uk.zoom.us/j/89473073451\n\nA recording of this talk is available for Un
 iversity members at the following link: https://www.cl.cam.ac.uk/seminars/
 wednesday/video/\n\nThis talk is being recorded. If you do not wish to be 
 seen in the recording\, please avoid sitting in the front three rows of se
 ats in the lecture theatre. Any questions asked will also be included in t
 he recording. The recording will be made available on the Department’s w
 ebpage
LOCATION:Lecture Theatre 1\, Computer Laboratory\, William Gates Building
END:VEVENT
END:VCALENDAR