An Analysis of Phenotypic Diversity in Multi-Solution Optimization

EasyChair Preprint 3286

Abstract

In optimization methods that return diverse solution sets, three interpretations of diversity can be distinguished: multi-objective optimization which searches diversity in objective space, multimodal optimization which tries spreading out the solutions in genetic space, and quality diversity which performs diversity maintenance in phenotypic space. We introduce niching methods that provide more flexibility to the analysis of diversity and a simple domain to compare and provide insights about the paradigms. We show that multiobjective optimization does not always produce much diversity, quality diversity is not sensitive to genetic neutrality and creates the most diverse set of solutions, and multimodal optimization produces higher fitness solutions. An autoencoder is used to discover phenotypic features automatically, producing an even more diverse solution set. Finally, we make recommendations about when to use which approach.

Keyphrases: Autoencoder, Evolutionary Computation, Multi-Solution Optimization, Solution Diversity, diversity, feature discovery, genetic neutrality, multi-objective optimization, multimodal optimization, phenotypic diversity, phenotypic feature, phenotypic niching space, quality diversity

@booklet{EasyChair:3286,
  author    = {Alexander Hagg and Mike Preuss and Alexander Asteroth and Thomas Bäck},
  title     = {An Analysis of Phenotypic Diversity in Multi-Solution Optimization},
  howpublished = {EasyChair Preprint 3286},
  year      = {EasyChair, 2020}}