Skip to main content

Lund University Publications

LUND UNIVERSITY LIBRARIES

A GENERAL FRAMEWORK FOR USER-GUIDED BAYESIAN OPTIMIZATION

Hvarfner, Carl LU ; Hutter, Frank and Nardi, Luigi LU (2024) 12th International Conference on Learning Representations, ICLR 2024
Abstract

The optimization of expensive-to-evaluate black-box functions is prevalent in various scientific disciplines. Bayesian optimization is an automatic, general and sample-efficient method to solve these problems with minimal knowledge of the underlying function dynamics. However, the ability of Bayesian optimization to incorporate prior knowledge or beliefs about the function at hand in order to accelerate the optimization is limited, which reduces its appeal for knowledgeable practitioners with tight budgets. To allow domain experts to customize the optimization routine, we propose ColaBO, the first Bayesian-principled framework for incorporating prior beliefs beyond the typical kernel structure, such as the likely location of the... (More)

The optimization of expensive-to-evaluate black-box functions is prevalent in various scientific disciplines. Bayesian optimization is an automatic, general and sample-efficient method to solve these problems with minimal knowledge of the underlying function dynamics. However, the ability of Bayesian optimization to incorporate prior knowledge or beliefs about the function at hand in order to accelerate the optimization is limited, which reduces its appeal for knowledgeable practitioners with tight budgets. To allow domain experts to customize the optimization routine, we propose ColaBO, the first Bayesian-principled framework for incorporating prior beliefs beyond the typical kernel structure, such as the likely location of the optimizer or the optimal value. The generality of ColaBO makes it applicable across different Monte Carlo acquisition functions and types of user beliefs. We empirically demonstrate ColaBO's ability to substantially accelerate optimization when the prior information is accurate, and to retain approximately default performance when it is misleading.

(Less)
Please use this url to cite or link to this publication:
author
; and
organization
publishing date
type
Contribution to conference
publication status
published
subject
conference name
12th International Conference on Learning Representations, ICLR 2024
conference location
Hybrid, Vienna, Austria
conference dates
2024-05-07 - 2024-05-11
external identifiers
  • scopus:85197024019
language
English
LU publication?
yes
id
f6627881-7118-4a14-afcd-9d27eb34a4f8
date added to LUP
2024-12-16 12:56:56
date last changed
2025-04-08 02:16:27
@misc{f6627881-7118-4a14-afcd-9d27eb34a4f8,
  abstract     = {{<p>The optimization of expensive-to-evaluate black-box functions is prevalent in various scientific disciplines. Bayesian optimization is an automatic, general and sample-efficient method to solve these problems with minimal knowledge of the underlying function dynamics. However, the ability of Bayesian optimization to incorporate prior knowledge or beliefs about the function at hand in order to accelerate the optimization is limited, which reduces its appeal for knowledgeable practitioners with tight budgets. To allow domain experts to customize the optimization routine, we propose ColaBO, the first Bayesian-principled framework for incorporating prior beliefs beyond the typical kernel structure, such as the likely location of the optimizer or the optimal value. The generality of ColaBO makes it applicable across different Monte Carlo acquisition functions and types of user beliefs. We empirically demonstrate ColaBO's ability to substantially accelerate optimization when the prior information is accurate, and to retain approximately default performance when it is misleading.</p>}},
  author       = {{Hvarfner, Carl and Hutter, Frank and Nardi, Luigi}},
  language     = {{eng}},
  title        = {{A GENERAL FRAMEWORK FOR USER-GUIDED BAYESIAN OPTIMIZATION}},
  year         = {{2024}},
}