BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:Symmetries via Duality - Halverson James
DTSTART:20210504T140000Z
DTEND:20210504T150000Z
UID:TALK160252@talks.cam.ac.uk
CONTACT:96082
DESCRIPTION:Many neural networks (NNs) are drawn from Gaussian processes i
 n an appropriate infinite N limit. At large-but-finite N\, the associated 
 non-Gaussian processes may be treated using techniques from quantum field 
 theory (QFT). One fundamental aspect of both NN and QFT distributions are 
 their symmetries\, which in this talk I will study via a duality between p
 arameter space and function space. As in physics\, via duality we can util
 ize one perspective to gain knowledge of the other. In this case\, we can 
 study NN correlation functions computed in parameter space to determine sy
 mmetries of the function space distribution\, even when it is not known. C
 entral results are obvious in the GP limit of IID-parameter neural network
 s\, but the duality allows us to extend the symmetry results to all values
  of N and some independence-breaking schemes.
LOCATION:https://us02web.zoom.us/j/4784296471?pwd=SE8vc1BvWldlZnc1YUNwK3Q1
 dHpodz09
END:VEVENT
END:VCALENDAR