Algorithms and SQ Lower Bounds for PAC Learning One-Hidden-Layer ReLU Networks

COLT, pp. 1514-1539, 2020.

Cited by: 0|Bibtex|Views34|Links
EI

Abstract:

We study the problem of PAC learning one-hidden-layer ReLU networks with $k$ hidden units on $\mathbb{R}^d$ under Gaussian marginals in the presence of additive label noise. For the case of positive coefficients, we give the first polynomial-time algorithm for this learning problem for $k$ up to $\tilde{O}(\sqrt{\log d})$. Previously, n...More

Code:

Data:

Your rating :
0

 

Tags
Comments