A PAC-Bayes bound for deterministic classifiers

09/06/2022

∙

by   Eugenio Clerico, et al.

∙

19

∙

share

We establish a disintegrated PAC-Bayesian bound, for classifiers that are trained via continuous-time (non-stochastic) gradient descent. Contrarily to what is standard in the PAC-Bayesian setting, our result applies to a training algorithm that is deterministic, conditioned on a random initialisation, without requiring any de-randomisation step. We provide a broad discussion of the main features of the bound that we propose, and we study analytically and empirically its behaviour on linear models, finding promising results.