Provable Accelerated Bayesian Optimization with Knowledge Transfer

We study how Bayesian optimization (BO) can be accelerated on a target task with historical knowledge transferred from related source tasks. Existing works on BO with knowledge transfer either do not have theoretical guarantees or achieve the same regret as BO in the non-transfer setting, Õ(√T γf), where T is the number of evaluations of the target function and γf denotes its information gain. In this paper, we propose the DeltaBO algorithm, in which a novel uncertainty-quantification approach is built on the difference function $δ$ between the source and target functions, which are allowed to belong to different reproducing kernel Hilbert spaces (RKHSs). Under mild assumptions, we prove that the regret of DeltaBO is of order Õ(√T (T/N + γ_δ)), where N denotes the number of evaluations from source tasks and typically N T. In many applications, source and target tasks are similar, which implies that γ_δ can be much smaller than γf. Empirical studies on both real-world hyperparameter tuning tasks and synthetic functions show that DeltaBO outperforms other baseline methods and support our theoretical claims.