Comparing Optimization Algorithms Through the Lens of Search Behavior Analysis

TL;DR

This paper evaluates the effectiveness of statistical tests, specifically the cross-match test, in comparing the search behaviors of numerous optimization algorithms to better understand their similarities and differences.

Contribution

It introduces a method for analyzing and comparing optimization algorithms based on their search behavior using statistical tests, addressing issues of obscured innovation.

Findings

Cross-match test effectively compares search behaviors

114 algorithms analyzed from MEALPY library

Identifies groups of algorithms with similar search patterns

Abstract

The field of numerical optimization has recently seen a surge in the development of "novel" metaheuristic algorithms, inspired by metaphors derived from natural or human-made processes, which have been widely criticized for obscuring meaningful innovations and failing to distinguish themselves from existing approaches. Aiming to address these concerns, we investigate the applicability of statistical tests for comparing algorithms based on their search behavior. We utilize the cross-match statistical test to compare multivariate distributions and assess the solutions produced by 114 algorithms from the MEALPY library. These findings are incorporated into an empirical analysis aiming to identify algorithms with similar search behaviors.