Projective Preferential Bayesian Optimization

Petrus Mikkola, Milica Todorović, Jari Järvi, Patrick Rinke, Samuel Kaski
Proceedings of the 37th International Conference on Machine Learning, PMLR 119:6884-6892, 2020.

Abstract

Bayesian optimization is an effective method for finding extrema of a black-box function. We propose a new type of Bayesian optimization for learning user preferences in high-dimensional spaces. The central assumption is that the underlying objective function cannot be evaluated directly, but instead a minimizer along a projection can be queried, which we call a projective preferential query. The form of the query allows for feedback that is natural for a human to give, and which enables interaction. This is demonstrated in a user experiment in which the user feedback comes in the form of optimal position and orientation of a molecule adsorbing to a surface. We demonstrate that our framework is able to find a global minimum of a high-dimensional black-box function, which is an infeasible task for existing preferential Bayesian optimization frameworks that are based on pairwise comparisons.

Cite this Paper


BibTeX
@InProceedings{pmlr-v119-mikkola20a, title = {Projective Preferential {B}ayesian Optimization}, author = {Mikkola, Petrus and Todorovi{\'c}, Milica and J{\"a}rvi, Jari and Rinke, Patrick and Kaski, Samuel}, booktitle = {Proceedings of the 37th International Conference on Machine Learning}, pages = {6884--6892}, year = {2020}, editor = {Hal Daumé III and Aarti Singh}, volume = {119}, series = {Proceedings of Machine Learning Research}, month = {13--18 Jul}, publisher = {PMLR}, pdf = {http://proceedings.mlr.press/v119/mikkola20a/mikkola20a.pdf}, url = { http://proceedings.mlr.press/v119/mikkola20a.html }, abstract = {Bayesian optimization is an effective method for finding extrema of a black-box function. We propose a new type of Bayesian optimization for learning user preferences in high-dimensional spaces. The central assumption is that the underlying objective function cannot be evaluated directly, but instead a minimizer along a projection can be queried, which we call a projective preferential query. The form of the query allows for feedback that is natural for a human to give, and which enables interaction. This is demonstrated in a user experiment in which the user feedback comes in the form of optimal position and orientation of a molecule adsorbing to a surface. We demonstrate that our framework is able to find a global minimum of a high-dimensional black-box function, which is an infeasible task for existing preferential Bayesian optimization frameworks that are based on pairwise comparisons.} }
Endnote
%0 Conference Paper %T Projective Preferential Bayesian Optimization %A Petrus Mikkola %A Milica Todorović %A Jari Järvi %A Patrick Rinke %A Samuel Kaski %B Proceedings of the 37th International Conference on Machine Learning %C Proceedings of Machine Learning Research %D 2020 %E Hal Daumé III %E Aarti Singh %F pmlr-v119-mikkola20a %I PMLR %P 6884--6892 %U http://proceedings.mlr.press/v119/mikkola20a.html %V 119 %X Bayesian optimization is an effective method for finding extrema of a black-box function. We propose a new type of Bayesian optimization for learning user preferences in high-dimensional spaces. The central assumption is that the underlying objective function cannot be evaluated directly, but instead a minimizer along a projection can be queried, which we call a projective preferential query. The form of the query allows for feedback that is natural for a human to give, and which enables interaction. This is demonstrated in a user experiment in which the user feedback comes in the form of optimal position and orientation of a molecule adsorbing to a surface. We demonstrate that our framework is able to find a global minimum of a high-dimensional black-box function, which is an infeasible task for existing preferential Bayesian optimization frameworks that are based on pairwise comparisons.
APA
Mikkola, P., Todorović, M., Järvi, J., Rinke, P. & Kaski, S.. (2020). Projective Preferential Bayesian Optimization. Proceedings of the 37th International Conference on Machine Learning, in Proceedings of Machine Learning Research 119:6884-6892 Available from http://proceedings.mlr.press/v119/mikkola20a.html .

Related Material