View in EDS

Intelligent decision-making for binary coverage: Unveiling the potential of the multi-armed bandit selector.

Saved in:
Bibliographic Details
Title: Intelligent decision-making for binary coverage: Unveiling the potential of the multi-armed bandit selector.
Authors: Becerra-Rozas, Marcelo1 (AUTHOR) marcelo.becerra.r@mail.pucv.cl, Lemus-Romani, José2 (AUTHOR) jose.lemus@uc.cl, Crawford, Broderick1 (AUTHOR) broderick.crawford@pucv.cl, Soto, Ricardo1 (AUTHOR) ricardo.soto@pucv.cl, Talbi, El-Ghazali3 (AUTHOR) el-ghazali.talbi@univ-lille.fr
Source: Expert Systems with Applications. Oct2024, Vol. 251, pN.PAG-N.PAG. 1p.
Subjects: Metaheuristic algorithms, Grey Wolf Optimizer algorithm, Evolutionary algorithms, Reinforcement learning, Machine learning, Decision making
Abstract: In this article, we propose the integration of a novel reinforcement learning technique into our generic and unified framework. This framework enables any continuous metaheuristic to operate in binary optimization, with the technique in question known as the Multi-Armed Bandit. Population-based metaheuristics comprise multiple individuals that cooperatively and globally explore the search space using their limited individual capabilities. Our framework allows these population-based metaheuristics to continue leveraging their original movements, designed for continuous optimization, once they are binary encoded. The generality of the framework has facilitated the instantiation of popular algorithms from the optimization, machine learning, and evolutionary computing communities. Furthermore, it permits the design of new and innovative optimization instances using various component strategies, reflecting the framework's modularity. The results comparing two statistical techniques and three hybridizations coming from Machine Learning, have shown to obtain a better performance with the metahuristics in Grey Wolf Optimizer and Whale Optimization Algorithm. [ABSTRACT FROM AUTHOR]
Copyright of Expert Systems with Applications is the property of Pergamon Press - An Imprint of Elsevier Science and its content may not be copied or emailed to multiple sites without the copyright holder's express written permission. Additionally, content may not be used with any artificial intelligence tools or machine learning technologies. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)
Database: Engineering Source
Description
Abstract:In this article, we propose the integration of a novel reinforcement learning technique into our generic and unified framework. This framework enables any continuous metaheuristic to operate in binary optimization, with the technique in question known as the Multi-Armed Bandit. Population-based metaheuristics comprise multiple individuals that cooperatively and globally explore the search space using their limited individual capabilities. Our framework allows these population-based metaheuristics to continue leveraging their original movements, designed for continuous optimization, once they are binary encoded. The generality of the framework has facilitated the instantiation of popular algorithms from the optimization, machine learning, and evolutionary computing communities. Furthermore, it permits the design of new and innovative optimization instances using various component strategies, reflecting the framework's modularity. The results comparing two statistical techniques and three hybridizations coming from Machine Learning, have shown to obtain a better performance with the metahuristics in Grey Wolf Optimizer and Whale Optimization Algorithm. [ABSTRACT FROM AUTHOR]
ISSN:09574174
DOI:10.1016/j.eswa.2024.124112