Accelerating Look-ahead in Bayesian Optimization: Multilevel Monte Carlo is All you Need

Fuente: arXiv
Enregistré dans:
Détails bibliographiques
Auteurs principaux: Yang, Shangda, Zankin, Vitaly, Balandat, Maximilian, Scherer, Stefan, Carlberg, Kevin, Walton, Neil, Law, Kody J. H.
Format: Preprint
Publié: 2024
Sujets:
Accès en ligne:
Tags: Ajouter un tag
Pas de tags, Soyez le premier à ajouter un tag!
_version_ 1866929443530866688
author Yang, Shangda
Zankin, Vitaly
Balandat, Maximilian
Scherer, Stefan
Carlberg, Kevin
Walton, Neil
Law, Kody J. H.
author_facet Yang, Shangda
Zankin, Vitaly
Balandat, Maximilian
Scherer, Stefan
Carlberg, Kevin
Walton, Neil
Law, Kody J. H.
contents We leverage multilevel Monte Carlo (MLMC) to improve the performance of multi-step look-ahead Bayesian optimization (BO) methods that involve nested expectations and maximizations. Often these expectations must be computed by Monte Carlo (MC). The complexity rate of naive MC degrades for nested operations, whereas MLMC is capable of achieving the canonical MC convergence rate for this type of problem, independently of dimension and without any smoothness assumptions. Our theoretical study focuses on the approximation improvements for twoand three-step look-ahead acquisition functions, but, as we discuss, the approach is generalizable in various ways, including beyond the context of BO. Our findings are verified numerically and the benefits of MLMC for BO are illustrated on several benchmark examples. Code is available at https://github.com/Shangda-Yang/MLMCBO .
format Preprint
id arxiv_https___arxiv_org_abs_2402_02111
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Accelerating Look-ahead in Bayesian Optimization: Multilevel Monte Carlo is All you Need
Yang, Shangda
Zankin, Vitaly
Balandat, Maximilian
Scherer, Stefan
Carlberg, Kevin
Walton, Neil
Law, Kody J. H.
Machine Learning
Optimization and Control
Probability
Computation
Methodology
We leverage multilevel Monte Carlo (MLMC) to improve the performance of multi-step look-ahead Bayesian optimization (BO) methods that involve nested expectations and maximizations. Often these expectations must be computed by Monte Carlo (MC). The complexity rate of naive MC degrades for nested operations, whereas MLMC is capable of achieving the canonical MC convergence rate for this type of problem, independently of dimension and without any smoothness assumptions. Our theoretical study focuses on the approximation improvements for twoand three-step look-ahead acquisition functions, but, as we discuss, the approach is generalizable in various ways, including beyond the context of BO. Our findings are verified numerically and the benefits of MLMC for BO are illustrated on several benchmark examples. Code is available at https://github.com/Shangda-Yang/MLMCBO .
title Accelerating Look-ahead in Bayesian Optimization: Multilevel Monte Carlo is All you Need
topic Machine Learning
Optimization and Control
Probability
Computation
Methodology
url https://arxiv.org/abs/2402.02111