What can phylogenetic metrics tell us about useful diversity in evolutionary algorithms?

TL;DR

This paper explores whether phylogenetic diversity metrics, which consider evolutionary history, can better predict the success of evolutionary algorithms compared to traditional diversity measures.

Contribution

It introduces the use of phylogenetic diversity metrics in evolutionary algorithms and evaluates their predictive power for long-term success.

Findings

Phylogenetic metrics often provide different information than traditional diversity metrics.

Phylogenetic diversity is a better predictor of evolutionary algorithm success.

Abstract

It is generally accepted that "diversity" is associated with success in evolutionary algorithms. However, diversity is a broad concept that can be measured and defined in a multitude of ways. To date, most evolutionary computation research has measured diversity using the richness and/or evenness of a particular genotypic or phenotypic property. While these metrics are informative, we hypothesize that other diversity metrics are more strongly predictive of success. Phylogenetic diversity metrics are a class of metrics popularly used in biology, which take into account the evolutionary history of a population. Here, we investigate the extent to which 1) these metrics provide different information than those traditionally used in evolutionary computation, and 2) these metrics better predict the long-term success of a run of evolutionary computation. We find that, in most cases, phylogenetic metrics behave meaningfully differently from other diversity metrics. Moreover, our results suggest that phylogenetic diversity is indeed a better predictor of success.