Interpreting Neural Networks as Majority Votes through the PAC-Bayesian Theory

Viallard, Paul; Emonet, Rémi; Germain, Pascal; Habrard, Amaury; Morvant, Emilie

Document type :

Autre communication scientifique (congrès sans actes - poster - séminaire...): Communication dans un congrès sans actes

Link :

https://lilloa.univ-lille.fr/handle/20.500.12210/29221

Title :

Interpreting Neural Networks as Majority Votes through the PAC-Bayesian Theory

Author(s) :

Viallard, Paul [Auteur]
Emonet, Rémi [Auteur]
Germain, Pascal [Auteur]
Habrard, Amaury [Auteur]
Morvant, Emilie [Auteur]

Conference title :

Workshop on Machine Learning with guarantees @ NeurIPS 2019

City :

Vancouver

Country :

Canada

Start date of the conference :

2019-12-14

Publication date :

2019

HAL domain(s) :

Statistiques [stat]/Machine Learning [stat.ML]

English abstract : [en]

We propose a PAC-Bayesian theoretical study of the two-phase learning procedure of a neural network introduced by Kawaguchi et al. (2017). In this procedure, a network is expressed as a weighted combination of all the paths ...
Show more >We propose a PAC-Bayesian theoretical study of the two-phase learning procedure of a neural network introduced by Kawaguchi et al. (2017). In this procedure, a network is expressed as a weighted combination of all the paths of the network (from the input layer to the output one), that we reformulate as a PAC-Bayesian majority vote. Starting from this observation, their learning procedure consists in (1) learning a "prior" network for fixing some parameters, then (2) learning a "posterior" network by only allowing a modification of the weights over the paths of the prior network. This allows us to derive a PAC-Bayesian generalization bound that involves the empirical individual risks of the paths (known as the Gibbs risk) and the empirical diversity between pairs of paths. Note that similarly to classical PAC-Bayesian bounds, our result involves a KL-divergence term between a "prior" network and the "posterior" network. We show that this term is computable by dynamic programming without assuming any distribution on the network weights.Show less >

Language :

Anglais

Audience :

Internationale

Popular science :

Non

Administrative institution(s) :

CNRS
Université de Lille

Collections :

METRICS : Evaluation des technologies de santé et des pratiques médicales - ULR 2694

Submission date :

2020-06-08T14:10:29Z

Interpreting Neural Networks as Majority ... BibTeX CSV Excel RIS

Interpreting Neural Networks as Majority ...

BibTeX

CSV

Excel

RIS