Exploring exploration in Bayesian 0ptimization
(2025)- Abstract
- A well-balanced exploration-exploitation trade-off is crucial for successful acquisition functions in Bayesian optimization. However, there is a lack of quantitative measures for exploration, making it difficult to analyze and compare different acquisition functions. This work introduces two novel approaches - observation traveling salesman distance and observation entropy - to quantify the exploration characteristics of acquisition functions based on their selected observations. Using these measures, we examine the explorative nature of several well-known acquisition functions across a diverse set of black-box problems, uncover links between exploration and empirical performance, and reveal new relationships among existing acquisition... (More)
- A well-balanced exploration-exploitation trade-off is crucial for successful acquisition functions in Bayesian optimization. However, there is a lack of quantitative measures for exploration, making it difficult to analyze and compare different acquisition functions. This work introduces two novel approaches - observation traveling salesman distance and observation entropy - to quantify the exploration characteristics of acquisition functions based on their selected observations. Using these measures, we examine the explorative nature of several well-known acquisition functions across a diverse set of black-box problems, uncover links between exploration and empirical performance, and reveal new relationships among existing acquisition functions. Beyond enabling a deeper understanding of acquisition functions, these measures also provide a foundation for guiding their design in a more principled and systematic manner. (Less)
Please use this url to cite or link to this publication:
https://lup.lub.lu.se/record/01d454d1-cafa-4035-bcc3-0f50c36c6db0
- author
- Papenmeier, Leonard
LU
; Cheng, Nuojin ; Becker, Stephen and Nardi, Luigi LU
- organization
- publishing date
- 2025
- type
- Chapter in Book/Report/Conference proceeding
- publication status
- in press
- subject
- host publication
- Forty-First Conference on Uncertainty in Artificial Intelligence
- pages
- 28 pages
- language
- English
- LU publication?
- yes
- id
- 01d454d1-cafa-4035-bcc3-0f50c36c6db0
- alternative location
- https://arxiv.org/abs/2502.08208
- date added to LUP
- 2025-05-07 15:17:44
- date last changed
- 2025-05-15 11:11:56
@inproceedings{01d454d1-cafa-4035-bcc3-0f50c36c6db0, abstract = {{A well-balanced exploration-exploitation trade-off is crucial for successful acquisition functions in Bayesian optimization. However, there is a lack of quantitative measures for exploration, making it difficult to analyze and compare different acquisition functions. This work introduces two novel approaches - observation traveling salesman distance and observation entropy - to quantify the exploration characteristics of acquisition functions based on their selected observations. Using these measures, we examine the explorative nature of several well-known acquisition functions across a diverse set of black-box problems, uncover links between exploration and empirical performance, and reveal new relationships among existing acquisition functions. Beyond enabling a deeper understanding of acquisition functions, these measures also provide a foundation for guiding their design in a more principled and systematic manner.}}, author = {{Papenmeier, Leonard and Cheng, Nuojin and Becker, Stephen and Nardi, Luigi}}, booktitle = {{Forty-First Conference on Uncertainty in Artificial Intelligence}}, language = {{eng}}, title = {{Exploring exploration in Bayesian 0ptimization}}, url = {{https://arxiv.org/abs/2502.08208}}, year = {{2025}}, }