Learning Relevant Contextual Variables Within Bayesian Optimization

Fuente: arXiv
Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: Martinelli, Julien, Bharti, Ayush, Tiihonen, Armi, John, S. T., Filstroff, Louis, Sloman, Sabina J., Rinke, Patrick, Kaski, Samuel
Format: Preprint
Veröffentlicht: 2023
Schlagworte:
Online-Zugang:
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
_version_ 1866913360025485312
author Martinelli, Julien
Bharti, Ayush
Tiihonen, Armi
John, S. T.
Filstroff, Louis
Sloman, Sabina J.
Rinke, Patrick
Kaski, Samuel
author_facet Martinelli, Julien
Bharti, Ayush
Tiihonen, Armi
John, S. T.
Filstroff, Louis
Sloman, Sabina J.
Rinke, Patrick
Kaski, Samuel
contents 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.
format Preprint
id arxiv_https___arxiv_org_abs_2305_14120
institution arXiv
publishDate 2023
record_format arxiv
spellingShingle Learning Relevant Contextual Variables Within Bayesian Optimization
Martinelli, Julien
Bharti, Ayush
Tiihonen, Armi
John, S. T.
Filstroff, Louis
Sloman, Sabina J.
Rinke, Patrick
Kaski, Samuel
Machine Learning
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.
title Learning Relevant Contextual Variables Within Bayesian Optimization
topic Machine Learning
url https://arxiv.org/abs/2305.14120