A learning automata based multiobjective hyper-heuristic

Metaheuristics, being tailored to each particular domain by experts, have been successfully applied to many computationally hard optimisation problems. However, once implemented, their application to a new problem domain or a slight change in the problem description would often require additional ex...

Full description

Bibliographic Details
Main Authors: Li, Wenwen, Özcan, Ender, John, Robert
Format: Article
Published: IEEE 2017
Subjects:
Online Access:https://eprints.nottingham.ac.uk/48692/
_version_ 1848797824036110336
author Li, Wenwen
Özcan, Ender
John, Robert
author_facet Li, Wenwen
Özcan, Ender
John, Robert
author_sort Li, Wenwen
building Nottingham Research Data Repository
collection Online Access
description Metaheuristics, being tailored to each particular domain by experts, have been successfully applied to many computationally hard optimisation problems. However, once implemented, their application to a new problem domain or a slight change in the problem description would often require additional expert intervention. There is a growing number of studies on reusable cross-domain search methodologies, such as, selection hyper-heuristics, which are applicable to problem instances from various domains, requiring minimal expert intervention or even none. This study introduces a new learning automata based selection hyper-heuristic controlling a set of multiobjective metaheuristics. The approach operates above three well-known multiobjective evolutionary algorithms and mixes them, exploiting the strengths of each algorithm. The performance and behaviour of two variants of the proposed selection hyper-heuristic, each utilising a different initialisation scheme are investigated across a range of unconstrained multiobjective mathematical benchmark functions from two different sets and the realworld problem of vehicle crashworthiness. The empirical results illustrate the effectiveness of our approach for cross-domain search, regardless of the initialisation scheme, on those problems when compared to each individual multiobjective algorithm. Moreover, both variants perform signicantly better than some previously proposed selection hyper-heuristics for multiobjective optimisation, thus signicantly enhancing the opportunities for improved multiobjective optimisation.
first_indexed 2025-11-14T20:10:00Z
format Article
id nottingham-48692
institution University of Nottingham Malaysia Campus
institution_category Local University
last_indexed 2025-11-14T20:10:00Z
publishDate 2017
publisher IEEE
recordtype eprints
repository_type Digital Repository
spelling nottingham-486922020-05-04T19:23:19Z https://eprints.nottingham.ac.uk/48692/ A learning automata based multiobjective hyper-heuristic Li, Wenwen Özcan, Ender John, Robert Metaheuristics, being tailored to each particular domain by experts, have been successfully applied to many computationally hard optimisation problems. However, once implemented, their application to a new problem domain or a slight change in the problem description would often require additional expert intervention. There is a growing number of studies on reusable cross-domain search methodologies, such as, selection hyper-heuristics, which are applicable to problem instances from various domains, requiring minimal expert intervention or even none. This study introduces a new learning automata based selection hyper-heuristic controlling a set of multiobjective metaheuristics. The approach operates above three well-known multiobjective evolutionary algorithms and mixes them, exploiting the strengths of each algorithm. The performance and behaviour of two variants of the proposed selection hyper-heuristic, each utilising a different initialisation scheme are investigated across a range of unconstrained multiobjective mathematical benchmark functions from two different sets and the realworld problem of vehicle crashworthiness. The empirical results illustrate the effectiveness of our approach for cross-domain search, regardless of the initialisation scheme, on those problems when compared to each individual multiobjective algorithm. Moreover, both variants perform signicantly better than some previously proposed selection hyper-heuristics for multiobjective optimisation, thus signicantly enhancing the opportunities for improved multiobjective optimisation. IEEE 2017-12-20 Article PeerReviewed Li, Wenwen, Özcan, Ender and John, Robert (2017) A learning automata based multiobjective hyper-heuristic. IEEE Transactions on Evolutionary Computation . ISSN 1089-778X Online learning Multiobjective optimisation Hyper-heuristics Evolutionary algorithms Operational research http://ieeexplore.ieee.org/document/8231198/ doi:10.1109/TEVC.2017.2785346 doi:10.1109/TEVC.2017.2785346
spellingShingle Online learning
Multiobjective optimisation
Hyper-heuristics
Evolutionary algorithms
Operational research
Li, Wenwen
Özcan, Ender
John, Robert
A learning automata based multiobjective hyper-heuristic
title A learning automata based multiobjective hyper-heuristic
title_full A learning automata based multiobjective hyper-heuristic
title_fullStr A learning automata based multiobjective hyper-heuristic
title_full_unstemmed A learning automata based multiobjective hyper-heuristic
title_short A learning automata based multiobjective hyper-heuristic
title_sort learning automata based multiobjective hyper-heuristic
topic Online learning
Multiobjective optimisation
Hyper-heuristics
Evolutionary algorithms
Operational research
url https://eprints.nottingham.ac.uk/48692/
https://eprints.nottingham.ac.uk/48692/
https://eprints.nottingham.ac.uk/48692/