Asynchronous Parallel Bayesian Optimisation via Thompson Sampling
This work addresses the challenge of parallelizing Bayesian optimization for expensive evaluations, such as in hyper-parameter tuning, offering a simpler and more effective method compared to prior approaches.
The paper tackles the problem of expensive function evaluations in Bayesian optimization by proposing asynchronous parallel Thompson sampling, showing that distributing n evaluations among M workers is essentially equivalent to sequential evaluations and achieves asymptotically lower regret under time constraints. Experimental results demonstrate that it outperforms existing parallel BO algorithms in simulations and hyper-parameter tuning for convolutional neural networks.
We design and analyse variations of the classical Thompson sampling (TS) procedure for Bayesian optimisation (BO) in settings where function evaluations are expensive, but can be performed in parallel. Our theoretical analysis shows that a direct application of the sequential Thompson sampling algorithm in either synchronous or asynchronous parallel settings yields a surprisingly powerful result: making $n$ evaluations distributed among $M$ workers is essentially equivalent to performing $n$ evaluations in sequence. Further, by modeling the time taken to complete a function evaluation, we show that, under a time constraint, asynchronously parallel TS achieves asymptotically lower regret than both the synchronous and sequential versions. These results are complemented by an experimental analysis, showing that asynchronous TS outperforms a suite of existing parallel BO algorithms in simulations and in a hyper-parameter tuning application in convolutional neural networks. In addition to these, the proposed procedure is conceptually and computationally much simpler than existing work for parallel BO.