Visualising Evolution History in Multi- and Many-Objective Optimisation

TL;DR

This paper presents a visualization framework for evolutionary algorithms applied to multi- and many-objective problems, enhancing understanding of complex problem landscapes and algorithm processes, especially when features are unknown or high-dimensional.

Contribution

It adapts an existing visualization technique to multi- and many-objective data, enabling better insight into the optimization process and problem features.

Findings

Effective visualization of evolutionary processes on benchmark problems

Improved understanding of problem landscapes in high-dimensional spaces

Facilitates analysis of unknown or complex problem features

Abstract

Evolutionary algorithms are widely used to solve optimisation problems. However, challenges of transparency arise in both visualising the processes of an optimiser operating through a problem and understanding the problem features produced from many-objective problems, where comprehending four or more spatial dimensions is difficult. This work considers the visualisation of a population as an optimisation process executes. We have adapted an existing visualisation technique to multi- and many-objective problem data, enabling a user to visualise the EA processes and identify specific problem characteristics and thus providing a greater understanding of the problem landscape. This is particularly valuable if the problem landscape is unknown, contains unknown features or is a many-objective problem. We have shown how using this framework is effective on a suite of multi- and many-objective benchmark test problems, optimising them with NSGA-II and NSGA-III.