On Performance Estimation in Automatic Algorithm Configuration
It addresses a foundational gap in automated parameter tuning, which is crucial for improving algorithm configuration methods in various domains.
The paper tackles the weak theoretical foundations of automatic algorithm configuration by proving the universal best performance estimator and establishing theoretical bounds on estimation error for finite and infinite configuration spaces, with verification through experiments on four scenarios.
Over the last decade, research on automated parameter tuning, often referred to as automatic algorithm configuration (AAC), has made significant progress. Although the usefulness of such tools has been widely recognized in real world applications, the theoretical foundations of AAC are still very weak. This paper addresses this gap by studying the performance estimation problem in AAC. More specifically, this paper first proves the universal best performance estimator in a practical setting, and then establishes theoretical bounds on the estimation error, i.e., the difference between the training performance and the true performance for a parameter configuration, considering finite and infinite configuration spaces respectively. These findings were verified in extensive experiments conducted on four algorithm configuration scenarios involving different problem domains. Moreover, insights for enhancing existing AAC methods are also identified.