Summer 2019

Frontiers of Efficient Neural-Network Learnability

Jun. 19, 2019 11:00 am12:30 pm

Add to Calendar


Adam Klivans


Room 116

What is the most expressive class of neural networks that can be learned, provably, in polynomial-time in a distribution-free setting? In this talk we will describe how to combine isotonic regression with kernel methods to give efficient algorithms for learning neural networks with two nonlinear layers.  We will touch upon relationships with recent work on SGD plus overparameterization.