Runtime Analyses of NSGA-III on Many-Objective Problems

TL;DR

This paper provides the first rigorous runtime analyses of NSGA-III on many-objective benchmark problems, offering insights into parameter settings for optimal performance as the number of objectives increases.

Contribution

It extends the theoretical understanding of NSGA-III to problems with more than three objectives, analyzing parameter scaling and performance guarantees.

Findings

Established parameter scaling rules for NSGA-III on many-objective problems.

Provided the first runtime bounds for NSGA-III on mLOTZ, mOMM, and mCOCZ benchmarks.

Demonstrated how to set reference points and population size for effective optimization.

Abstract

NSGA-II and NSGA-III are two of the most popular evolutionary multi-objective algorithms used in practice. While NSGA-II is used for few objectives such as 2 and 3, NSGA-III is designed to deal with a larger number of objectives. In a recent breakthrough, Wietheger and Doerr (IJCAI 2023) gave the first runtime analysis for NSGA-III on the 3-objective OneMinMax problem, showing that this state-of-the-art algorithm can be analyzed rigorously. We advance this new line of research by presenting the first runtime analyses of NSGA-III on the popular many-objective benchmark problems mLOTZ, mOMM, and mCOCZ, for an arbitrary constant number $m$ of objectives. Our analysis provides ways to set the important parameters of the algorithm: the number of reference points and the population size, so that a good performance can be guaranteed. We show how these parameters should be scaled with the problem dimension, the number of objectives and the fitness range. To our knowledge, these are the first runtime analyses for NSGA-III for more than 3 objectives.