Discovering the Elite Hypervolume by Leveraging Interspecies Correlation

TL;DR

This paper introduces the concept of an elite hypervolume in evolutionary algorithms, proposing metrics and a new variation operator that leverages interspecies correlations to improve the MAP-Elites algorithm's efficiency across diverse problems.

Contribution

It defines the elite hypervolume concept, introduces metrics for its characterization, and proposes a directional variation operator that accelerates MAP-Elites by exploiting inter-elites correlations.

Findings

The directional variation operator improves convergence speed.

The elite hypervolume can be characterized by genotypic spread and similarity.

Effective across different problem domains.

Abstract

Evolution has produced an astonishing diversity of species, each filling a different niche. Algorithms like MAP-Elites mimic this divergent evolutionary process to find a set of behaviorally diverse but high-performing solutions, called the elites. Our key insight is that species in nature often share a surprisingly large part of their genome, in spite of occupying very different niches; similarly, the elites are likely to be concentrated in a specific "elite hypervolume" whose shape is defined by their common features. In this paper, we first introduce the elite hypervolume concept and propose two metrics to characterize it: the genotypic spread and the genotypic similarity. We then introduce a new variation operator, called "directional variation", that exploits interspecies (or inter-elites) correlations to accelerate the MAP-Elites algorithm. We demonstrate the effectiveness of this operator in three problems (a toy function, a redundant robotic arm, and a hexapod robot).