Predicting Effective Control Parameters for Differential Evolution using Cluster Analysis of Objective Function Features

TL;DR

This paper presents a method that uses simple objective function features and cluster analysis to predict effective control parameters for differential evolution, outperforming existing adaptive techniques in benchmark tests.

Contribution

It introduces a novel approach combining objective function features and cluster analysis to improve control parameter selection in differential evolution.

Findings

Method outperforms state-of-the-art techniques in benchmarks.

Simple features effectively predict control parameters.

Most computation is done offline, with significant online performance gains.

Abstract

A methodology is introduced which uses three simple objective function features to predict effective control parameters for differential evolution. This is achieved using cluster analysis techniques to classify objective functions using these features. Information on prior performance of various control parameters for each classification is then used to determine which control parameters to use in future optimisations. Our approach is compared to state-of-the-art adaptive and non-adaptive techniques. Two accepted bench mark suites are used to compare performance and in all cases we show that the improvement resulting from our approach is statistically significant. The majority of the computational effort of this methodology is performed off-line, however even when taking into account the additional on-line cost our approach outperforms other adaptive techniques. We also investigate the key tuning parameters of our methodology, such as number of clusters, which further support the finding that the simple features selected are predictors of effective control parameters. The findings presented in this paper are significant because they show that simple to calculate features of objective functions can help to select control parameters for optimisation algorithms. This can have an immediate positive impact on the application of these optimisation algorithms on real world problems, where it is often difficult to select effective control parameters.