Minimum Regret Search for Single- and Multi-Task Optimization
This work addresses optimization efficiency for machine learning and robotics, but it is incremental as it builds on existing information-theoretic methods like entropy search.
The authors tackled the problem of Bayesian optimization by proposing a new acquisition function called minimum regret search (MRS), which aims to minimize expected simple regret for optimal recommendations, and found that MRS reduces high-regret outliers compared to entropy search (ES) in synthetic and robotic control tasks.
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.