BORE: Bayesian Optimization by Density-Ratio Estimation

Louis C Tiao, Aaron Klein, Matthias W Seeger, Edwin V. Bonilla, Cedric Archambeau, Fabio Ramos
Proceedings of the 38th International Conference on Machine Learning, PMLR 139:10289-10300, 2021.

Abstract

Bayesian optimization (BO) is among the most effective and widely-used blackbox optimization methods. BO proposes solutions according to an explore-exploit trade-off criterion encoded in an acquisition function, many of which are computed from the posterior predictive of a probabilistic surrogate model. Prevalent among these is the expected improvement (EI). The need to ensure analytical tractability of the predictive often poses limitations that can hinder the efficiency and applicability of BO. In this paper, we cast the computation of EI as a binary classification problem, building on the link between class-probability estimation and density-ratio estimation, and the lesser-known link between density-ratios and EI. By circumventing the tractability constraints, this reformulation provides numerous advantages, not least in terms of expressiveness, versatility, and scalability.

Cite this Paper


BibTeX
@InProceedings{pmlr-v139-tiao21a, title = {BORE: Bayesian Optimization by Density-Ratio Estimation}, author = {Tiao, Louis C and Klein, Aaron and Seeger, Matthias W and Bonilla, Edwin V. and Archambeau, Cedric and Ramos, Fabio}, booktitle = {Proceedings of the 38th International Conference on Machine Learning}, pages = {10289--10300}, year = {2021}, editor = {Meila, Marina and Zhang, Tong}, volume = {139}, series = {Proceedings of Machine Learning Research}, month = {18--24 Jul}, publisher = {PMLR}, pdf = {http://proceedings.mlr.press/v139/tiao21a/tiao21a.pdf}, url = {http://proceedings.mlr.press/v139/tiao21a.html}, abstract = {Bayesian optimization (BO) is among the most effective and widely-used blackbox optimization methods. BO proposes solutions according to an explore-exploit trade-off criterion encoded in an acquisition function, many of which are computed from the posterior predictive of a probabilistic surrogate model. Prevalent among these is the expected improvement (EI). The need to ensure analytical tractability of the predictive often poses limitations that can hinder the efficiency and applicability of BO. In this paper, we cast the computation of EI as a binary classification problem, building on the link between class-probability estimation and density-ratio estimation, and the lesser-known link between density-ratios and EI. By circumventing the tractability constraints, this reformulation provides numerous advantages, not least in terms of expressiveness, versatility, and scalability.} }
Endnote
%0 Conference Paper %T BORE: Bayesian Optimization by Density-Ratio Estimation %A Louis C Tiao %A Aaron Klein %A Matthias W Seeger %A Edwin V. Bonilla %A Cedric Archambeau %A Fabio Ramos %B Proceedings of the 38th International Conference on Machine Learning %C Proceedings of Machine Learning Research %D 2021 %E Marina Meila %E Tong Zhang %F pmlr-v139-tiao21a %I PMLR %P 10289--10300 %U http://proceedings.mlr.press/v139/tiao21a.html %V 139 %X Bayesian optimization (BO) is among the most effective and widely-used blackbox optimization methods. BO proposes solutions according to an explore-exploit trade-off criterion encoded in an acquisition function, many of which are computed from the posterior predictive of a probabilistic surrogate model. Prevalent among these is the expected improvement (EI). The need to ensure analytical tractability of the predictive often poses limitations that can hinder the efficiency and applicability of BO. In this paper, we cast the computation of EI as a binary classification problem, building on the link between class-probability estimation and density-ratio estimation, and the lesser-known link between density-ratios and EI. By circumventing the tractability constraints, this reformulation provides numerous advantages, not least in terms of expressiveness, versatility, and scalability.
APA
Tiao, L.C., Klein, A., Seeger, M.W., Bonilla, E.V., Archambeau, C. & Ramos, F.. (2021). BORE: Bayesian Optimization by Density-Ratio Estimation. Proceedings of the 38th International Conference on Machine Learning, in Proceedings of Machine Learning Research 139:10289-10300 Available from http://proceedings.mlr.press/v139/tiao21a.html.

Related Material