Enregistré dans:
Détails bibliographiques
Auteurs principaux: Wei, Qiyu, Wang, Haowei, Cao, Zirui, Wang, Songhao, Allmendinger, Richard, Álvarez, Mauricio A
Format: Preprint
Publié: 2025
Sujets:
Accès en ligne:https://arxiv.org/abs/2504.07742
Tags: Ajouter un tag
Pas de tags, Soyez le premier à ajouter un tag!
_version_ 1866912640740098048
author Wei, Qiyu
Wang, Haowei
Cao, Zirui
Wang, Songhao
Allmendinger, Richard
Álvarez, Mauricio A
author_facet Wei, Qiyu
Wang, Haowei
Cao, Zirui
Wang, Songhao
Allmendinger, Richard
Álvarez, Mauricio A
contents Bayesian optimization (BO) is an effective technique for black-box optimization. However, its applicability is typically limited to moderate-budget problems due to the cubic complexity of fitting the Gaussian process (GP) surrogate model. In large-budget scenarios, directly employing the standard GP model faces significant challenges in computational time and resource requirements. In this paper, we propose a novel approach, gradient-based sample selection Bayesian Optimization (GSSBO), to enhance the computational efficiency of BO. The GP model is constructed on a selected set of samples instead of the whole dataset. These samples are selected by leveraging gradient information to remove redundancy while preserving diversity and representativeness. We provide a theoretical analysis of the gradient-based sample selection strategy and obtain explicit sublinear regret bounds for our proposed framework. Extensive experiments on synthetic and real-world tasks demonstrate that our approach significantly reduces the computational cost of GP fitting in BO while maintaining optimization performance comparable to baseline methods.
format Preprint
id arxiv_https___arxiv_org_abs_2504_07742
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Gradient-based Sample Selection for Faster Bayesian Optimization
Wei, Qiyu
Wang, Haowei
Cao, Zirui
Wang, Songhao
Allmendinger, Richard
Álvarez, Mauricio A
Machine Learning
Bayesian optimization (BO) is an effective technique for black-box optimization. However, its applicability is typically limited to moderate-budget problems due to the cubic complexity of fitting the Gaussian process (GP) surrogate model. In large-budget scenarios, directly employing the standard GP model faces significant challenges in computational time and resource requirements. In this paper, we propose a novel approach, gradient-based sample selection Bayesian Optimization (GSSBO), to enhance the computational efficiency of BO. The GP model is constructed on a selected set of samples instead of the whole dataset. These samples are selected by leveraging gradient information to remove redundancy while preserving diversity and representativeness. We provide a theoretical analysis of the gradient-based sample selection strategy and obtain explicit sublinear regret bounds for our proposed framework. Extensive experiments on synthetic and real-world tasks demonstrate that our approach significantly reduces the computational cost of GP fitting in BO while maintaining optimization performance comparable to baseline methods.
title Gradient-based Sample Selection for Faster Bayesian Optimization
topic Machine Learning
url https://arxiv.org/abs/2504.07742