Evaluation of Genotypic Diversity Measurements Exploited in Real-Coded Representation

TL;DR

This paper evaluates four genotypic diversity measures using a new validation framework based on three key requirements, revealing that none fully satisfy all criteria, highlighting the complexity of accurately assessing population diversity in evolutionary algorithms.

Contribution

The study introduces a validation framework with three specific requirements for genotypic diversity measures and applies it to assess four existing GDMs.

Findings

None of the evaluated GDMs fully meet all validation criteria.

Proper assessment of population diversity remains a challenging task.

The proposed framework helps identify limitations of current GDMs.

Abstract

Numerous genotypic diversity measures (GDMs) are available in the literature to assess the convergence status of an evolutionary algorithm (EA) or describe its search behavior. In a recent study, the authors of this paper drew attention to the need for a GDM validation framework. In response, this study proposes three requirements (monotonicity in individual varieties, twinning, and monotonicity in distance) that can clearly portray any GDMs. These diversity requirements are analysed by means of controlled population arrangements. In this paper four GDMs are evaluated with the proposed validation framework. The results confirm that properly evaluating population diversity is a rather difficult task, as none of the analysed GDMs complies with all the diversity requirements.