[ugrads] SIAM Series of Research Talks
ycao33 at hawk.iit.edu
Thu Feb 28 13:38:02 CST 2019
Next Wednesday we will have our 2nd research talk by Prof Dixon. The talk is about the Machine Learning. We will offer lunch. See you there.
Time: 12:50, Wednesday, 6th March
Speaker: Matthew Dixon
Title: Voronoi Neural Networks
We introduce a multi-layer perceptron (MLP) which explicitly tessellates the domain with a Voronoi diagram rather than learns the partitions through optimization. The primary advantage of such a network is that is equivalent to b-spine interpolation, where the degree of the b-spline matches that of the activation functions. Under Lipschitz continuity of the target function, we present various bounds on the global approximation error as a function of the number of hidden units. Numerical experiments are presented to demonstrate the convergence properties. Finally, some important future directions are discussed, such as the effect of multiple hidden layers on the approximation error.
Yue Cao, PHD candidate
IIT SIAM Student Chapter President
Department of Applied Mathematics
Illinois of Institute of Technology
Office: John T. Rettaliata Engineering Center, Room 232
Email: ycao33 at hawk.iit.edu
-------------- next part --------------
An HTML attachment was scrubbed...
More information about the ugrads