English

Information in Infinite Ensembles of Infinitely-Wide Neural Networks

Machine Learning 2022-11-08 v3 Information Theory math.IT Machine Learning

Abstract

In this preliminary work, we study the generalization properties of infinite ensembles of infinitely-wide neural networks. Amazingly, this model family admits tractable calculations for many information-theoretic quantities. We report analytical and empirical investigations in the search for signals that correlate with generalization.

Keywords

Cite

@article{arxiv.1911.09189,
  title  = {Information in Infinite Ensembles of Infinitely-Wide Neural Networks},
  author = {Ravid Shwartz-Ziv and Alexander A. Alemi},
  journal= {arXiv preprint arXiv:1911.09189},
  year   = {2022}
}

Comments

2nd Symposium on Advances in Approximate Bayesian Inference, 2019