This repository includes the code associated to the paper:
The codes uses the implementation of Lipschitz-continuous neural networks provided by Anil et al., 2019, as well as parts of the implementation of probabilistic neural networks provided by Perez-Ortiz et al., 2021.
Citation:
@inproceedings{mbacke2023pac,
title={PAC-Bayesian Generalization Guarantees for Adversarial Generative Models},
author={Mbacke, Sokhna Diarra and Clerc, Florence and Germain, Pascal},
booktitle={International Conference on Machine Learning},
year={2023},
organization={PMLR}
}