Salvato in:
Dettagli Bibliografici
Autori principali: Bardou, Anthony, Gonon, Antoine, Ahadinia, Aryan, Thiran, Patrick
Natura: Preprint
Pubblicazione: 2025
Soggetti:
Accesso online:https://arxiv.org/abs/2509.25051
Tags: Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
_version_ 1866908566325035008
author Bardou, Anthony
Gonon, Antoine
Ahadinia, Aryan
Thiran, Patrick
author_facet Bardou, Anthony
Gonon, Antoine
Ahadinia, Aryan
Thiran, Patrick
contents Bayesian Optimization (BO) is a powerful framework for optimizing noisy, expensive-to-evaluate black-box functions. When the objective exhibits invariances under a group action, exploiting these symmetries can substantially improve BO efficiency. While using maximum similarity across group orbits has long been considered in other domains, the fact that the max kernel is not positive semidefinite (PSD) has prevented its use in BO. In this work, we revisit this idea by considering a PSD projection of the max kernel. Compared to existing invariant (and non-invariant) kernels, we show it achieves significantly lower regret on both synthetic and real-world BO benchmarks, without increasing computational complexity.
format Preprint
id arxiv_https___arxiv_org_abs_2509_25051
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Symmetry-Aware Bayesian Optimization via Max Kernels
Bardou, Anthony
Gonon, Antoine
Ahadinia, Aryan
Thiran, Patrick
Machine Learning
Bayesian Optimization (BO) is a powerful framework for optimizing noisy, expensive-to-evaluate black-box functions. When the objective exhibits invariances under a group action, exploiting these symmetries can substantially improve BO efficiency. While using maximum similarity across group orbits has long been considered in other domains, the fact that the max kernel is not positive semidefinite (PSD) has prevented its use in BO. In this work, we revisit this idea by considering a PSD projection of the max kernel. Compared to existing invariant (and non-invariant) kernels, we show it achieves significantly lower regret on both synthetic and real-world BO benchmarks, without increasing computational complexity.
title Symmetry-Aware Bayesian Optimization via Max Kernels
topic Machine Learning
url https://arxiv.org/abs/2509.25051