Multi-View Majority Vote Learning Algorithms: Direct Minimization of PAC-Bayesian Bounds

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Hennequin, Mehdi, Zitouni, Abdelkrim, Benabdeslem, Khalid, Elghazel, Haytham, Gaci, Yacine
Format: Preprint
Published: 2024
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866911207865188352
author Hennequin, Mehdi
Zitouni, Abdelkrim
Benabdeslem, Khalid
Elghazel, Haytham
Gaci, Yacine
author_facet Hennequin, Mehdi
Zitouni, Abdelkrim
Benabdeslem, Khalid
Elghazel, Haytham
Gaci, Yacine
contents The PAC-Bayesian framework has significantly advanced the understanding of statistical learning, particularly for majority voting methods. Despite its successes, its application to multi-view learning -- a setting with multiple complementary data representations -- remains underexplored. In this work, we extend PAC-Bayesian theory to multi-view learning, introducing novel generalization bounds based on Rényi divergence. These bounds provide an alternative to traditional Kullback-Leibler divergence-based counterparts, leveraging the flexibility of Rényi divergence. Furthermore, we propose first- and second-order oracle PAC-Bayesian bounds and extend the C-bound to multi-view settings. To bridge theory and practice, we design efficient self-bounding optimization algorithms that align with our theoretical results.
format Preprint
id arxiv_https___arxiv_org_abs_2411_06276
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Multi-View Majority Vote Learning Algorithms: Direct Minimization of PAC-Bayesian Bounds
Hennequin, Mehdi
Zitouni, Abdelkrim
Benabdeslem, Khalid
Elghazel, Haytham
Gaci, Yacine
Machine Learning
Artificial Intelligence
The PAC-Bayesian framework has significantly advanced the understanding of statistical learning, particularly for majority voting methods. Despite its successes, its application to multi-view learning -- a setting with multiple complementary data representations -- remains underexplored. In this work, we extend PAC-Bayesian theory to multi-view learning, introducing novel generalization bounds based on Rényi divergence. These bounds provide an alternative to traditional Kullback-Leibler divergence-based counterparts, leveraging the flexibility of Rényi divergence. Furthermore, we propose first- and second-order oracle PAC-Bayesian bounds and extend the C-bound to multi-view settings. To bridge theory and practice, we design efficient self-bounding optimization algorithms that align with our theoretical results.
title Multi-View Majority Vote Learning Algorithms: Direct Minimization of PAC-Bayesian Bounds
topic Machine Learning
Artificial Intelligence
url https://arxiv.org/abs/2411.06276