Abstract
Coevolving memetic algorithms are a family of metaheuristic search algorithms in which a rule-based representation of local search (LS) is coadapted alongside candidate solutions within a hybrid evolutionary system. Simple versions of these systems have been shown to outperform other nonadaptive memetic and evolutionary algorithms on a range of problems. This paper presents a rationale for such systems and places them in the context of other recent work on adaptive memetic algorithms. It then proposes a general structure within which a population of LS algorithms can be evolved in tandem with the solutions to which they are applied. Previous research started with a simple self-adaptive system before moving on to more complex models. Results showed that the algorithm was able to discover and exploit certain forms of structure and regularities within the problems. This "metalearning" of problem features provided a means of creating highly scalable algorithms. This work is briefly reviewed to highlight some of the important findings and behaviors exhibited. Based on this analysis, new results are then presented from systems with more flexible representations, which, again, show significant improvements. Finally, the current state of, and future directions for, research in this area is discussed.
Highlights
T HE performance benefits which can be achieved by hybridising Evolutionary Algorithms (EAs) with Local Search (LS) operators, so-called Memetic Algorithms (MAs), have been well documented across a wide range of problem domains such as optimisation of combinatorial, nonstationary and multi-objective problems
On these problems there was a clear benefit to using adaptive neighbourhood local search, since the RandCOMA algorithm found the optimum on every run, the Success Rate metric did not provide conclusive evidence that learning was taking place
The premise that evolutionary optimisation algorithms can be improved by incorporating an appropriate local search mechanism is widely accepted, but it is increasingly recognised that the choice of move operator, and neighbourhood function, used in the local search is crucial to delivering success
Summary
T HE performance benefits which can be achieved by hybridising Evolutionary Algorithms (EAs) with Local Search (LS) operators, so-called Memetic Algorithms (MAs), have been well documented across a wide range of problem domains such as optimisation of combinatorial, nonstationary and multi-objective problems (see [1] for a review, [2] for a collection of recent algorithmic and theoretical work and [3] for a comprehensive bibliography). Previous papers have reported initial results from a system within which the definitions of Local Search operators applied within the MA may be changed during the course of optimisation This was named the COevolving Memetic Algorithm (COMA). This paper places the COMA framework in the context of algorithmic advances by other authors, and reviews the progression of research and results from the initial simple systems [4]–[6] to the more complex, truly co-evolutionary systems tested in [7] In those investigations the emphasis had been placed on evolving the rule-based neighbourhood definition but leaving much of the rest of Local Search algorithm fixed.
Sign-in/Register to view highlights & summary Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callMore From: IEEE Transactions on Systems, Man and Cybernetics, Part B (Cybernetics)
Disclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
