A Greedy Optimization Approach for High-Dimensional Bayesian Optimization

The standard Bayesian optimization algorithm encounters the curse of dimensionality in high-dimensional problems. The difficulty of optimizing the acquisition function increases, resulting in low sampling efficiency and high computational costs. In this work, we propose a simple and efficient method to solve this problem. We propose a greedy Bayesian optimization (GBO) algorithm, which uses greedy approach to optimize the acquisition function. Instead of optimizing the acquisition function in the original high-dimensional space, we optimize it group by group. We conduct numerical experiments and compare our proposed method with the standard Bayesian optimization algorithm and four state-of-the-art high- dimensional Bayesian optimization algorithms to validate the effectiveness of our proposed method. The results indicate that our proposed method demonstrates more competitive performance.