@inproceedings{discovery10138174,
       publisher = {Advances in Neural Information Processing Systems (NeurIPS 2021)},
            year = {2021},
       booktitle = {Proceedings of the 35th Conference on Neural Information Processing Systems (NeurIPS 2021)},
           title = {Learning Stochastic Majority Votes by Minimizing a PAC-Bayes Generalization Bound},
         journal = {CoRR},
            note = {This version is the version of record. For information on re-use, please refer to the publisher's terms and conditions.},
             url = {https://proceedings.neurips.cc/paper/2021/hash/0415740eaa4d9decbc8da001d3fd805f-Abstract.html},
          author = {Zantedeschi, V and Viallard, P and Morvant, E and Emonet, R and Habrard, A and Germain, P and Guedj, B},
        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.}
}