Learning relevant contextual variables within Bayesian optimization
No Thumbnail Available
Access rights
openAccess
URL
Journal Title
Journal ISSN
Volume Title
A4 Artikkeli konferenssijulkaisussa
This publication is imported from Aalto University research portal.
View publication in the Research portal (opens in new window)
Other link related to publication (opens in new window)
View publication in the Research portal (opens in new window)
Other link related to publication (opens in new window)
Date
2024
Major/Subject
Mcode
Degree programme
Language
en
Pages
Series
Proceedings of the Fortieth Conference on Uncertainty in Artificial Intelligence, Volume 244, pp. 2450-2470, Proceedings of Machine Learning Research ; Volume 244
Abstract
Contextual Bayesian Optimization (CBO) efficiently optimizes black-box functions with respect to design variables, while simultaneously integrating contextual information regarding the environment, such as experimental conditions. However, the relevance of contextual variables is not necessarily known beforehand. Moreover, contextual variables can sometimes be optimized themselves at an additional cost, a setting overlooked by current CBO algorithms. Cost-sensitive CBO would simply include optimizable contextual variables as part of the design variables based on their cost. Instead, we adaptively select a subset of contextual variables to include in the optimization, based on the trade-off between their relevance and the additional cost incurred by optimizing them compared to leaving them to be determined by the environment. We learn the relevance of contextual variables by sensitivity analysis of the posterior surrogate model while minimizing the cost of optimization by leveraging recent developments on early stopping for BO. We empirically evaluate our proposed Sensitivity-Analysis-Driven Contextual BO (SADCBO) method against alternatives on both synthetic and real-world experiments, together with extensive ablation studies, and demonstrate a consistent improvement across examples.Description
| openaire: EC/HE/101059891/EU//ACCELERATE-PER
Keywords
Other note
Citation
Martinelli, J, Bharti, A, Tiihonen, A, John, S T, Filstroff, L, Sloman, S J, Rinke, P & Kaski, S 2024, Learning relevant contextual variables within Bayesian optimization . in Proceedings of the Fortieth Conference on Uncertainty in Artificial Intelligence . vol. 244, Proceedings of Machine Learning Research, vol. 244, JMLR, pp. 2450-2470, Conference on Uncertainty in Artificial Intelligence, Barcelona, Spain, 15/07/2024 . < https://arxiv.org/abs/2305.14120 >