Guardado en:
Detalles Bibliográficos
Autores principales: Hu, Shouri, Li, Jiawei, Cai, Zhibo
Formato: Preprint
Publicado: 2024
Materias:
Acceso en línea:https://arxiv.org/abs/2403.05425
Etiquetas: Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
_version_ 1866914708192231424
author Hu, Shouri
Li, Jiawei
Cai, Zhibo
author_facet Hu, Shouri
Li, Jiawei
Cai, Zhibo
contents Bayesian optimization (BO) has shown impressive results in a variety of applications within low-to-moderate dimensional Euclidean spaces. However, extending BO to high-dimensional settings remains a significant challenge. We address this challenge by proposing a two-step optimization framework. Initially, we identify the effective dimension reduction (EDR) subspace for the objective function using the minimum average variance estimation (MAVE) method. Subsequently, we construct a Gaussian process model within this EDR subspace and optimize it using the expected improvement criterion. Our algorithm offers the flexibility to operate these steps either concurrently or in sequence. In the sequential approach, we meticulously balance the exploration-exploitation trade-off by distributing the sampling budget between subspace estimation and function optimization, and the convergence rate of our algorithm in high-dimensional contexts has been established. Numerical experiments validate the efficacy of our method in challenging scenarios.
format Preprint
id arxiv_https___arxiv_org_abs_2403_05425
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle An Adaptive Dimension Reduction Estimation Method for High-dimensional Bayesian Optimization
Hu, Shouri
Li, Jiawei
Cai, Zhibo
Machine Learning
Methodology
Bayesian optimization (BO) has shown impressive results in a variety of applications within low-to-moderate dimensional Euclidean spaces. However, extending BO to high-dimensional settings remains a significant challenge. We address this challenge by proposing a two-step optimization framework. Initially, we identify the effective dimension reduction (EDR) subspace for the objective function using the minimum average variance estimation (MAVE) method. Subsequently, we construct a Gaussian process model within this EDR subspace and optimize it using the expected improvement criterion. Our algorithm offers the flexibility to operate these steps either concurrently or in sequence. In the sequential approach, we meticulously balance the exploration-exploitation trade-off by distributing the sampling budget between subspace estimation and function optimization, and the convergence rate of our algorithm in high-dimensional contexts has been established. Numerical experiments validate the efficacy of our method in challenging scenarios.
title An Adaptive Dimension Reduction Estimation Method for High-dimensional Bayesian Optimization
topic Machine Learning
Methodology
url https://arxiv.org/abs/2403.05425