.This paper presents a new approach for batch Bayesian optimization (BO) called Thompson Sampling-Regret to Sigma Ratio directed sampling (TS-RSR), where we sample a new batch of actions by minimizing a TS approximation of a regret to uncertainty ratio. Our sampling objective is to coordinate the actions chosen in each batch in a way that minimizes redundancy between points while focusing on points with high predictive means or high uncertainty. Theoretically, we provide rigorous convergence guarantees on our algorithm's regret, and numerically we demonstrate that our method attains state-of-the-art performance on a range of challenging synthetic and realistic test functions, where it outperforms several competitive benchmark batch BO algorithms.KeywordsBayesian optimizationbatch Bayesian optimizationinformation-directed samplingMSC codes68Q2568R1068U05
Ren et al. (Fri,) studied this question.