Learning Stochastic Majority Votes by Minimizing a PAC-Bayes Generalization Bound
Abstract
We investigate a stochastic counterpart of majority votes over finite ensembles of classifiers, and study its generalization properties. While our approach holds for arbitrary distributions, we instantiate it with Dirichlet distributions: this allows for a closed-form and differentiable expression for the expected risk, which then turns the generalization bound into a tractable training objective. The resulting stochastic majority vote learning algorithm achieves state-of-the-art accuracy and benefits from (non-vacuous) tight generalization bounds, in a series of numerical experiments when compared to competing algorithms which also minimize PAC-Bayes objectives -- both with uninformed (data-independent) and informed (data-dependent) priors.
Cite
@article{arxiv.2106.12535,
title = {Learning Stochastic Majority Votes by Minimizing a PAC-Bayes Generalization Bound},
author = {Valentina Zantedeschi and Paul Viallard and Emilie Morvant and Rémi Emonet and Amaury Habrard and Pascal Germain and Benjamin Guedj},
journal= {arXiv preprint arXiv:2106.12535},
year = {2021}
}