ORTHOBO: Orthogonal Bayesian Hyperparameter Optimization

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Schröder, Maresa, Janetzky, Pascal, Klar, Michael, Feuerriegel, Stefan
Format: Preprint
Published: 2026
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866915989658009600
author Schröder, Maresa
Janetzky, Pascal
Klar, Michael
Feuerriegel, Stefan
author_facet Schröder, Maresa
Janetzky, Pascal
Klar, Michael
Feuerriegel, Stefan
contents Bayesian optimization is widely used for hyperparameter optimization when model evaluations are expensive; however, noisy acquisition estimates can lead to unstable decisions. We identify acquisition estimation noise as a failure mode that was previously overlooked: even when the surrogate model and acquisition target are correctly specified, finite-sample Monte Carlo error can perturb acquisition values. This can, in turn, flip candidate rankings and lead to suboptimal BO decisions. As a remedy, we aim at variance reduction and propose an orthogonal acquisition estimator that subtracts an optimally weighted score-function control variate, which yields an acquisition residual orthogonal to posterior score directions and which thus reduces Monte Carlo variance. We further introduce OrthoBO: a Bayesian optimization framework that combines our orthogonal acquisition estimator with ensemble surrogates and an outer log transformation. We show theoretically that our estimator preserves the target, leads to variance reduction, and improves pairwise ranking stability. We further verify the theoretical properties of OrthoBO through numerical experiments where our framework reduces acquisition estimation variance, stabilizes candidate rankings, and achieves strong performance. We also demonstrate the downstream utility of OrthoBO in hyperparameter optimization for neural network training and fine-tuning.
format Preprint
id arxiv_https___arxiv_org_abs_2605_06454
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle ORTHOBO: Orthogonal Bayesian Hyperparameter Optimization
Schröder, Maresa
Janetzky, Pascal
Klar, Michael
Feuerriegel, Stefan
Machine Learning
Artificial Intelligence
Bayesian optimization is widely used for hyperparameter optimization when model evaluations are expensive; however, noisy acquisition estimates can lead to unstable decisions. We identify acquisition estimation noise as a failure mode that was previously overlooked: even when the surrogate model and acquisition target are correctly specified, finite-sample Monte Carlo error can perturb acquisition values. This can, in turn, flip candidate rankings and lead to suboptimal BO decisions. As a remedy, we aim at variance reduction and propose an orthogonal acquisition estimator that subtracts an optimally weighted score-function control variate, which yields an acquisition residual orthogonal to posterior score directions and which thus reduces Monte Carlo variance. We further introduce OrthoBO: a Bayesian optimization framework that combines our orthogonal acquisition estimator with ensemble surrogates and an outer log transformation. We show theoretically that our estimator preserves the target, leads to variance reduction, and improves pairwise ranking stability. We further verify the theoretical properties of OrthoBO through numerical experiments where our framework reduces acquisition estimation variance, stabilizes candidate rankings, and achieves strong performance. We also demonstrate the downstream utility of OrthoBO in hyperparameter optimization for neural network training and fine-tuning.
title ORTHOBO: Orthogonal Bayesian Hyperparameter Optimization
topic Machine Learning
Artificial Intelligence
url https://arxiv.org/abs/2605.06454