<< May 2020 >>
 
 >>
 >>
 >>
 >>
 >>
 >>
Sun Mon Tue Wed Thu Fri Sat
   1   2 
 3   4   5   6   7   8   9 
 10   11   12   13   14   15   16 
 17   18   19   20   21   22   23 
 24   25   26   27   28   29   30 
 31   
 
Add an Event Edit This Event

Events at Physics

<< Fall 2019 Spring 2020 Summer 2020 >>
Subscribe to receive email announcements of events
Physics ∩ ML Seminar
Natural Graph Networks
Date: Wednesday, May 6th
Time: 11:00 am
Place: Please register for this online event: http://physicsmeetsml.org
Speaker: Taco Cohen, Qualcomm AI Research
Abstract: Message passing algorithms are the core of most neural networks that process information on graphs. Conventionally, such methods are invariant under permutation of the messages and hence forget how the information flows through the network. Analyzing the local symmetries of the graph, we show that a more general message passing network can in fact be sensitive the flow of information by using different kernels on different edges. This leads to an equivariant message passing algorithm that is more expressive than conventional invariant message passing, overcoming fundamental limitations of the latter. We derive the weight sharing and kernel constraints by modelling the symmetries using elementary category theory and show that equivariant kernels are “just” natural transformations between two functors. This general formulation, which we call Natural Networks, gives a unified theory to model many distinct forms of equivariant neural networks.
Host: Gary Shiu
Add this event to your calendar

©2013 Board of Regents of the University of Wisconsin System