Abstract

We derive information-theoretic generalization bounds for supervised learning algorithms based on the information contained in predictions rather than in the output of the training algorithm. These bounds improve over the existing information-theoretic bounds, are applicable to a wider range of algorithms, and solve two key challenges:(a) they give meaningful results for deterministic algorithms and (b) they are significantly easier to estimate. We show experimentally that the proposed bounds closely follow the generalization gap in practical scenarios for deep learning.

Date

December 6, 2021

Authors

Hrayr Harutyunyan, Maxim Raginsky, Greg Ver Steeg, Aram Galstyan

Journal

Advances in Neural Information Processing Systems

Volume

34

Pages

24670-24682