A Stratified Analysis of Bayesian Optimization Methods

TL;DR

This paper introduces a stratified empirical analysis framework for Bayesian optimization methods, using specific metrics and ranking to evaluate performance across different test function categories that mimic hyperparameter tuning challenges.

Contribution

It proposes a novel performance comparison approach with metrics and ranking mechanisms tailored for stratified test functions in Bayesian optimization.

Findings

Metrics effectively differentiate Bayesian methods across function types

Ranking provides clear performance summaries within strata

Test functions mimic real hyperparameter optimization complexities

Abstract

Empirical analysis serves as an important complement to theoretical analysis for studying practical Bayesian optimization. Often empirical insights expose strengths and weaknesses inaccessible to theoretical analysis. We define two metrics for comparing the performance of Bayesian optimization methods and propose a ranking mechanism for summarizing performance within various genres or strata of test functions. These test functions serve to mimic the complexity of hyperparameter optimization problems, the most prominent application of Bayesian optimization, but with a closed form which allows for rapid evaluation and more predictable behavior. This offers a flexible and efficient way to investigate functions with specific properties of interest, such as oscillatory behavior or an optimum on the domain boundary.