Cost-aware Stopping for Bayesian Optimization

Qian Xie, Linda Cai, Alexander Terenin, Peter I. Frazier, Ziv Scully

Submission Type: Short paper

Tldr: We propose a cost-aware stopping rule for Bayesian optimization that are theoretically grounded, free of heuristic tuning, and consistently achieve competitive cost-adjusted simple regret on empirical AutoML hyperparameter-tuning tasks.

Submission Number: 18