Minimum Regret Search for Single- and Multi-Task Optimization

Author name not available (Why is that?)

Publication date: 2 February 2016

Abstract: We propose minimum regret search (MRS), a novel acquisition function for Bayesian optimization. MRS bears similarities with information-theoretic approaches such as entropy search (ES). However, while ES aims in each query at maximizing the information gain with respect to the global maximum, MRS aims at minimizing the expected simple regret of its ultimate recommendation for the optimum. While empirically ES and MRS perform similar in most of the cases, MRS produces fewer outliers with high simple regret than ES. We provide empirical results both for a synthetic single-task optimization problem as well as for a simulated multi-task robotic control problem.

Has companion code repository: https://github.com/jmetzen/bayesian_optimization

This page was built for publication: Minimum Regret Search for Single- and Multi-Task Optimization

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q6270051)