LABCAT: Locally adaptive Bayesian optimization using principal-component-aligned trust regions

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Visser, E., van Daalen, C. E., Schoeman, J. C.
Format: Preprint
Published: 2023
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866909224276066304
author Visser, E.
van Daalen, C. E.
Schoeman, J. C.
author_facet Visser, E.
van Daalen, C. E.
Schoeman, J. C.
contents Bayesian optimization (BO) is a popular method for optimizing expensive black-box functions. BO has several well-documented shortcomings, including computational slowdown with longer optimization runs, poor suitability for non-stationary or ill-conditioned objective functions, and poor convergence characteristics. Several algorithms have been proposed that incorporate local strategies, such as trust regions, into BO to mitigate these limitations; however, none address all of them satisfactorily. To address these shortcomings, we propose the LABCAT algorithm, which extends trust-region-based BO by adding a rotation aligning the trust region with the weighted principal components and an adaptive rescaling strategy based on the length-scales of a local Gaussian process surrogate model with automatic relevance determination. Through extensive numerical experiments using a set of synthetic test functions and the well-known COCO benchmarking software, we show that the LABCAT algorithm outperforms several state-of-the-art BO and other black-box optimization algorithms.
format Preprint
id arxiv_https___arxiv_org_abs_2311_11328
institution arXiv
publishDate 2023
record_format arxiv
spellingShingle LABCAT: Locally adaptive Bayesian optimization using principal-component-aligned trust regions
Visser, E.
van Daalen, C. E.
Schoeman, J. C.
Machine Learning
Bayesian optimization (BO) is a popular method for optimizing expensive black-box functions. BO has several well-documented shortcomings, including computational slowdown with longer optimization runs, poor suitability for non-stationary or ill-conditioned objective functions, and poor convergence characteristics. Several algorithms have been proposed that incorporate local strategies, such as trust regions, into BO to mitigate these limitations; however, none address all of them satisfactorily. To address these shortcomings, we propose the LABCAT algorithm, which extends trust-region-based BO by adding a rotation aligning the trust region with the weighted principal components and an adaptive rescaling strategy based on the length-scales of a local Gaussian process surrogate model with automatic relevance determination. Through extensive numerical experiments using a set of synthetic test functions and the well-known COCO benchmarking software, we show that the LABCAT algorithm outperforms several state-of-the-art BO and other black-box optimization algorithms.
title LABCAT: Locally adaptive Bayesian optimization using principal-component-aligned trust regions
topic Machine Learning
url https://arxiv.org/abs/2311.11328