Dichotomize and Generalize: PAC-Bayesian Binary Activated Deep Neural Networks - Inria - Institut national de recherche en sciences et technologies du numérique Access content directly
Conference Papers Year : 2019

Dichotomize and Generalize: PAC-Bayesian Binary Activated Deep Neural Networks

Abstract

We present a comprehensive study of multilayer neural networks with binary activation, relying on the PAC-Bayesian theory. Our contributions are twofold: (i) we develop an end-to-end framework to train a binary activated deep neural network, overcoming the fact that binary activation function is non-differentiable; (ii) we provide nonvacuous PAC-Bayesian generalization bounds for binary activated deep neural networks. Noteworthy, our results are obtained by minimizing the expected loss of an architecture-dependent aggregation of binary activated deep neural networks. The performance of our approach is assessed on a thorough numerical experiment protocol on real-life datasets.
Fichier principal
Vignette du fichier
main.pdf (3.51 Mo) Télécharger le fichier
Origin : Files produced by the author(s)
Loading...

Dates and versions

hal-02139432 , version 1 (24-05-2019)
hal-02139432 , version 2 (29-05-2019)

Identifiers

  • HAL Id : hal-02139432 , version 2

Cite

Gaël Letarte, Pascal Germain, Benjamin Guedj, François Laviolette. Dichotomize and Generalize: PAC-Bayesian Binary Activated Deep Neural Networks. NeurIPS 2019 - Thirty-third Conference on Neural Information Processing Systems, Dec 2019, Vancouver, Canada. ⟨hal-02139432v2⟩
135 View
190 Download

Share

Gmail Facebook X LinkedIn More