The limit of infinite width allows for substantial simplifications in the analytical study of over-parameterised neural networks. With a suitable random initialisation, an extremely large network exhibits an approximately Gaussian behaviour. In the present work, we establish a similar result for a simple stochastic architecture whose parameters are random variables, holding both before and during training. The explicit evaluation of the output distribution allows for a PAC-Bayesian training procedure that directly optimises the generalisation bound. For a large but finite-width network, we show empirically on MNIST that this training approach can outperform standard PAC-Bayesian methods.
@article{arxiv.2106.09798,
title = {Wide stochastic networks: Gaussian limit and PAC-Bayesian training},
author = {Eugenio Clerico and George Deligiannidis and Arnaud Doucet},
journal= {arXiv preprint arXiv:2106.09798},
year = {2023}
}