ML Anthology

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

Behnam Neyshabur, Srinadh Bhojanapalli, Nathan Srebro
ICLR 2018
/iclr/2018/neyshabur2018iclr-pacbayesian/

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.

PDF ICLR Semantic Scholar

Cite

Text

Neyshabur et al. "A PAC-Bayesian Approach to Spectrally-Normalized Margin Bounds for Neural Networks." International Conference on Learning Representations, 2018.

Markdown

[Neyshabur et al. "A PAC-Bayesian Approach to Spectrally-Normalized Margin Bounds for Neural Networks." International Conference on Learning Representations, 2018.](https://mlanthology.org/iclr/2018/neyshabur2018iclr-pacbayesian/)

BibTeX

@inproceedings{neyshabur2018iclr-pacbayesian,
  title     = {{A PAC-Bayesian Approach to Spectrally-Normalized Margin Bounds for Neural Networks}},
  author    = {Neyshabur, Behnam and Bhojanapalli, Srinadh and Srebro, Nathan},
  booktitle = {International Conference on Learning Representations},
  year      = {2018},
  url       = {https://mlanthology.org/iclr/2018/neyshabur2018iclr-pacbayesian/}
}