Non-elitist Evolutionary Multi-objective Optimizers Revisited

TL;DR

This study challenges the conventional belief that elitist EMOAs always outperform non-elitist ones by demonstrating that non-elitist EMOAs can perform competitively on bi-objective problems with many variables when using an unbounded archive.

Contribution

The paper provides a comprehensive evaluation of non-elitist EMOAs, showing their competitive performance and analyzing their properties in bi-objective continuous optimization.

Findings

Non-elitist EMOAs perform well with certain crossover methods.

Unbounded external archives enhance non-elitist EMOA performance.

Non-elitist strategies can be effective on high-dimensional problems.

Abstract

Since around 2000, it has been considered that elitist evolutionary multi-objective optimization algorithms (EMOAs) always outperform non-elitist EMOAs. This paper revisits the performance of non-elitist EMOAs for bi-objective continuous optimization when using an unbounded external archive. This paper examines the performance of EMOAs with two elitist and one non-elitist environmental selections. The performance of EMOAs is evaluated on the bi-objective BBOB problem suite provided by the COCO platform. In contrast to conventional wisdom, results show that non-elitist EMOAs with particular crossover methods perform significantly well on the bi-objective BBOB problems with many decision variables when using the unbounded external archive. This paper also analyzes the properties of the non-elitist selection.