Reactive Search strategies using Reinforcement Learning, local search algorithms and Variable Neighborhood Search》.pdf

Expert Systems with Applications 41 (2014) 4939–4949 Contents lists available at ScienceDirect Expert Systems with Applications journal homepag e: /locate/eswa Review Reactive Search strategies using Reinforcement Learning, local search algorithms and Variable Neighborhood Search João Paulo Queiroz dos Santos a,⇑, Jorge Dantas de Melo b, Adrião Dória Duarte Neto b, Daniel Aloise b a Federal Institute of Education, Science and Technology of Rio Grande do Norte, Campus Natal-Zona Norte, Natal, RN, Brazil b Federal University of Rio Grande do Norte, Department of Computer Engineering and Automation, Natal, RN, Brazil a r t i c l e i n f o a b s t r a c t Keywords: Optimization techniques known as metaheuristics have been applied successfully to solve different prob- Reactive Search lems, in which their development is characterized by the appropriate selection of parameters (values) for Reinforcement Learning its execution. Where the adjustment of a parameter is required, this parameter will be tested until viable Local search results are obtained. Normally, such adjustments are made by the developer deploying the metaheuristic. Variable Neighborhood Search The quality of the results of a test instance [The term instance is used to refer to the assignment of values Combinatorial optimization to the input variables of a problem.] will not be transferred to the instances that were not tested yet and