Model and Feature Diversity for Bayesian Neural Networks in Mutual Learning

Fuente: arXiv
Enregistré dans:
Détails bibliographiques
Auteurs principaux: Pham, Cuong, Nguyen, Cuong C., Le, Trung, Phung, Dinh, Carneiro, Gustavo, Do, Thanh-Toan
Format: Preprint
Publié: 2024
Sujets:
Accès en ligne:
Tags: Ajouter un tag
Pas de tags, Soyez le premier à ajouter un tag!
_version_ 1866917711741714432
author Pham, Cuong
Nguyen, Cuong C.
Le, Trung
Phung, Dinh
Carneiro, Gustavo
Do, Thanh-Toan
author_facet Pham, Cuong
Nguyen, Cuong C.
Le, Trung
Phung, Dinh
Carneiro, Gustavo
Do, Thanh-Toan
contents Bayesian Neural Networks (BNNs) offer probability distributions for model parameters, enabling uncertainty quantification in predictions. However, they often underperform compared to deterministic neural networks. Utilizing mutual learning can effectively enhance the performance of peer BNNs. In this paper, we propose a novel approach to improve BNNs performance through deep mutual learning. The proposed approaches aim to increase diversity in both network parameter distributions and feature distributions, promoting peer networks to acquire distinct features that capture different characteristics of the input, which enhances the effectiveness of mutual learning. Experimental results demonstrate significant improvements in the classification accuracy, negative log-likelihood, and expected calibration error when compared to traditional mutual learning for BNNs.
format Preprint
id arxiv_https___arxiv_org_abs_2407_02721
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Model and Feature Diversity for Bayesian Neural Networks in Mutual Learning
Pham, Cuong
Nguyen, Cuong C.
Le, Trung
Phung, Dinh
Carneiro, Gustavo
Do, Thanh-Toan
Machine Learning
Computer Vision and Pattern Recognition
Bayesian Neural Networks (BNNs) offer probability distributions for model parameters, enabling uncertainty quantification in predictions. However, they often underperform compared to deterministic neural networks. Utilizing mutual learning can effectively enhance the performance of peer BNNs. In this paper, we propose a novel approach to improve BNNs performance through deep mutual learning. The proposed approaches aim to increase diversity in both network parameter distributions and feature distributions, promoting peer networks to acquire distinct features that capture different characteristics of the input, which enhances the effectiveness of mutual learning. Experimental results demonstrate significant improvements in the classification accuracy, negative log-likelihood, and expected calibration error when compared to traditional mutual learning for BNNs.
title Model and Feature Diversity for Bayesian Neural Networks in Mutual Learning
topic Machine Learning
Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2407.02721