Domain Knowledge Injection in Bayesian Search for New Materials
This work addresses the problem of biased knowledge injection in Bayesian optimization for materials science researchers, representing an incremental improvement over existing methods.
The authors tackled the challenge of incorporating domain knowledge into Bayesian optimization without introducing bias, proposing DKIBO, which uses a deterministic surrogate model to enhance the Gaussian process, and demonstrated its utility in a materials design task with empirical validation.
In this paper we propose DKIBO, a Bayesian optimization (BO) algorithm that accommodates domain knowledge to tune exploration in the search space. Bayesian optimization has recently emerged as a sample-efficient optimizer for many intractable scientific problems. While various existing BO frameworks allow the input of prior beliefs to accelerate the search by narrowing down the space, incorporating such knowledge is not always straightforward and can often introduce bias and lead to poor performance. Here we propose a simple approach to incorporate structural knowledge in the acquisition function by utilizing an additional deterministic surrogate model to enrich the approximation power of the Gaussian process. This is suitably chosen according to structural information of the problem at hand and acts a corrective term towards a better-informed sampling. We empirically demonstrate the practical utility of the proposed method by successfully injecting domain knowledge in a materials design task. We further validate our method's performance on different experimental settings and ablation analyses.