A PAC-Bayesian Approach to Spectrally-Normalized Margin Bounds for Neural Networks

ICLR, Volume abs/1707.09564, 2018.

Cited by: 227|Views83
EI

Abstract:

We present a generalization bound for feedforward neural networks in terms of the product of the spectral norm of the layers and the Frobenius norm of the weights. The generalization bound is derived using a PAC-Bayes analysis.

Code:

Data:

Your rating :
0

 

Tags
Comments