A hyper-heuristic approach to automated generation of mutation operators for evolutionary programming

Evolutionary programming can solve black-box function optimisation problems by evolving a population of numerical vectors. The variation component in the evolutionary process is supplied by a mutation operator, which is typically a Gaussian, Cauchy, or Lévy probability distribution. In this paper, w...

Full description

Bibliographic Details
Main Authors: Hong, Libin, Drake, John H., Woodward, John R., Özcan, Ender
Format: Article
Published: Elsevier 2017
Subjects:
Online Access:https://eprints.nottingham.ac.uk/47294/
_version_ 1848797509871206400
author Hong, Libin
Drake, John H.
Woodward, John R.
Özcan, Ender
author_facet Hong, Libin
Drake, John H.
Woodward, John R.
Özcan, Ender
author_sort Hong, Libin
building Nottingham Research Data Repository
collection Online Access
description Evolutionary programming can solve black-box function optimisation problems by evolving a population of numerical vectors. The variation component in the evolutionary process is supplied by a mutation operator, which is typically a Gaussian, Cauchy, or Lévy probability distribution. In this paper, we use genetic programming to automatically generate mutation operators for an evolutionary programming system, testing the proposed approach over a set of function classes, which represent a source of functions. The empirical results over a set of benchmark function classes illustrate that genetic programming can evolve mutation operators which generalise well from the training set to the test set on each function class. The proposed method is able to outperform existing human designed mutation operators with statistical significance in most cases, with competitive results observed for the rest.
first_indexed 2025-11-14T20:05:01Z
format Article
id nottingham-47294
institution University of Nottingham Malaysia Campus
institution_category Local University
last_indexed 2025-11-14T20:05:01Z
publishDate 2017
publisher Elsevier
recordtype eprints
repository_type Digital Repository
spelling nottingham-472942020-05-04T19:12:42Z https://eprints.nottingham.ac.uk/47294/ A hyper-heuristic approach to automated generation of mutation operators for evolutionary programming Hong, Libin Drake, John H. Woodward, John R. Özcan, Ender Evolutionary programming can solve black-box function optimisation problems by evolving a population of numerical vectors. The variation component in the evolutionary process is supplied by a mutation operator, which is typically a Gaussian, Cauchy, or Lévy probability distribution. In this paper, we use genetic programming to automatically generate mutation operators for an evolutionary programming system, testing the proposed approach over a set of function classes, which represent a source of functions. The empirical results over a set of benchmark function classes illustrate that genetic programming can evolve mutation operators which generalise well from the training set to the test set on each function class. The proposed method is able to outperform existing human designed mutation operators with statistical significance in most cases, with competitive results observed for the rest. Elsevier 2017-10-14 Article PeerReviewed Hong, Libin, Drake, John H., Woodward, John R. and Özcan, Ender (2017) A hyper-heuristic approach to automated generation of mutation operators for evolutionary programming. Applied Soft Computing, 62 . pp. 162-175. ISSN 1872-9681 Evolutionary Programming; Genetic Programming; Automatic Design; Hyper-heuristics; Continuous Optimization http://www.sciencedirect.com/science/article/pii/S1568494617306051 doi:10.1016/j.asoc.2017.10.002 doi:10.1016/j.asoc.2017.10.002
spellingShingle Evolutionary Programming; Genetic Programming; Automatic Design; Hyper-heuristics; Continuous Optimization
Hong, Libin
Drake, John H.
Woodward, John R.
Özcan, Ender
A hyper-heuristic approach to automated generation of mutation operators for evolutionary programming
title A hyper-heuristic approach to automated generation of mutation operators for evolutionary programming
title_full A hyper-heuristic approach to automated generation of mutation operators for evolutionary programming
title_fullStr A hyper-heuristic approach to automated generation of mutation operators for evolutionary programming
title_full_unstemmed A hyper-heuristic approach to automated generation of mutation operators for evolutionary programming
title_short A hyper-heuristic approach to automated generation of mutation operators for evolutionary programming
title_sort hyper-heuristic approach to automated generation of mutation operators for evolutionary programming
topic Evolutionary Programming; Genetic Programming; Automatic Design; Hyper-heuristics; Continuous Optimization
url https://eprints.nottingham.ac.uk/47294/
https://eprints.nottingham.ac.uk/47294/
https://eprints.nottingham.ac.uk/47294/