Beyond Regrets: Geometric Metrics for Bayesian Optimization

arXiv:2401.01981v2 Announce Type: replace
Abstract: Bayesian optimization is a principled optimization strategy for a black-box objective function. It shows its effectiveness in a wide variety of real-world applications such as scientific discovery and experimental design. In general, the performance of Bayesian optimization is reported through regret-based metrics such as instantaneous, simple, and cumulative regrets. These metrics only rely on function evaluations, so that they do not consider geometric relationships between query points and global solutions, or query points themselves. Notably, …

abstract applications arxiv bayesian beyond box cs.lg design discovery experimental function general metrics optimization optimization strategy performance scientific discovery shows simple stat.ml strategy through type world