Skip to content
conftrace_
2018 ICLR ICLR 2018

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

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.

🚀 Conference Pioneer - ICLR 2018

Authors

Behnam Neyshabur, Srinadh Bhojanapalli, Nathan Srebro

Topics

Machine Learning > Optimization & Theory > Learning Theory Machine Learning > Optimization & Theory > Statistical Learning Deep Learning > Architectures > Neural Networks

Keywords

spectral norm generalization bound margin bound frobenius norm neural network
Download PDF

More from ICLR 2018

All ICLR 2018 papers →
Learning From Noisy Singly-labeled Data
Understanding Short-Horizon Bias in Stochastic Meta-Optimization
A Scalable Laplace Approximation for Neural Networks
Deep Learning as a Mixed Convex-Combinatorial Optimization Problem
Leveraging Grammar and Reinforcement Learning for Neural Program Synthesis
Unbiased Online Recurrent Optimization