Discovering Many Diverse Solutions with Bayesian Optimization

Natalie Maus, Kaiwen Wu, David Eriksson, Jacob Gardner
Proceedings of The 26th International Conference on Artificial Intelligence and Statistics, PMLR 206:1779-1798, 2023.

Abstract

Bayesian optimization (BO) is a popular approach for sample-efficient optimization of black-box objective functions. While BO has been successfully applied to a wide range of scientific applications, traditional approaches to single-objective BO only seek to find a single best solution. This can be a significant limitation in situations where solutions may later turn out to be intractable, for example, a designed molecule may turn out to later violate constraints that can only be evaluated after the optimization process has concluded. To address this issue, we propose rank-ordered Bayesian Optimization with trustregions (ROBOT) which aims to find a portfolio of high-performing solutions that are diverse according to a user-specified diversity measure. We evaluate ROBOT on several real-world applications and show that it can discover large sets of high-performing diverse solutions while requiring few additional function evaluations compared to finding a single best solution.

Cite this Paper

BibTeX
@InProceedings{pmlr-v206-maus23a, title = {Discovering Many Diverse Solutions with Bayesian Optimization}, author = {Maus, Natalie and Wu, Kaiwen and Eriksson, David and Gardner, Jacob}, booktitle = {Proceedings of The 26th International Conference on Artificial Intelligence and Statistics}, pages = {1779--1798}, year = {2023}, editor = {Ruiz, Francisco and Dy, Jennifer and van de Meent, Jan-Willem}, volume = {206}, series = {Proceedings of Machine Learning Research}, month = {25--27 Apr}, publisher = {PMLR}, pdf = {https://proceedings.mlr.press/v206/maus23a/maus23a.pdf}, url = {https://proceedings.mlr.press/v206/maus23a.html}, abstract = {Bayesian optimization (BO) is a popular approach for sample-efficient optimization of black-box objective functions. While BO has been successfully applied to a wide range of scientific applications, traditional approaches to single-objective BO only seek to find a single best solution. This can be a significant limitation in situations where solutions may later turn out to be intractable, for example, a designed molecule may turn out to later violate constraints that can only be evaluated after the optimization process has concluded. To address this issue, we propose rank-ordered Bayesian Optimization with trustregions (ROBOT) which aims to find a portfolio of high-performing solutions that are diverse according to a user-specified diversity measure. We evaluate ROBOT on several real-world applications and show that it can discover large sets of high-performing diverse solutions while requiring few additional function evaluations compared to finding a single best solution.} }
Endnote
%0 Conference Paper %T Discovering Many Diverse Solutions with Bayesian Optimization %A Natalie Maus %A Kaiwen Wu %A David Eriksson %A Jacob Gardner %B Proceedings of The 26th International Conference on Artificial Intelligence and Statistics %C Proceedings of Machine Learning Research %D 2023 %E Francisco Ruiz %E Jennifer Dy %E Jan-Willem van de Meent %F pmlr-v206-maus23a %I PMLR %P 1779--1798 %U https://proceedings.mlr.press/v206/maus23a.html %V 206 %X Bayesian optimization (BO) is a popular approach for sample-efficient optimization of black-box objective functions. While BO has been successfully applied to a wide range of scientific applications, traditional approaches to single-objective BO only seek to find a single best solution. This can be a significant limitation in situations where solutions may later turn out to be intractable, for example, a designed molecule may turn out to later violate constraints that can only be evaluated after the optimization process has concluded. To address this issue, we propose rank-ordered Bayesian Optimization with trustregions (ROBOT) which aims to find a portfolio of high-performing solutions that are diverse according to a user-specified diversity measure. We evaluate ROBOT on several real-world applications and show that it can discover large sets of high-performing diverse solutions while requiring few additional function evaluations compared to finding a single best solution.
APA
Maus, N., Wu, K., Eriksson, D. & Gardner, J.. (2023). Discovering Many Diverse Solutions with Bayesian Optimization. Proceedings of The 26th International Conference on Artificial Intelligence and Statistics, in Proceedings of Machine Learning Research 206:1779-1798 Available from https://proceedings.mlr.press/v206/maus23a.html.

Related Material