Minimum Regret Search for Single- and Multi-Task Optimization
In International Conference on Machine Learning, (ICML), 19.6.-24.6.2016, New York, o.A., Jun/2016.
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 out- liers 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.