Weaving of Metaheuristics with Cooperative Parallelism
We propose PHYSH (Parallel HYbridization for Simple Heu-ristics), a framework to ease the design and implementation of hybridmetaheuristics via cooperative parallelism. With this framework, theuser only needs encode each of the desired metaheuristics and may relyon PHYSH for parallelization, coopera...
| Autor principal: | |
|---|---|
| Outros Autores: | , , |
| Formato: | article |
| Idioma: | eng |
| Publicado em: |
2018
|
| Texto completo: | http://hdl.handle.net/10174/23637 |
| País: | Portugal |
| Oai: | oai:dspace.uevora.pt:10174/23637 |
| Resumo: | We propose PHYSH (Parallel HYbridization for Simple Heu-ristics), a framework to ease the design and implementation of hybridmetaheuristics via cooperative parallelism. With this framework, theuser only needs encode each of the desired metaheuristics and may relyon PHYSH for parallelization, cooperation and hybridization. PHYSHsupports the combination of population-based and single-solution meta-heuristics and enables the user to control the tradeoff between intensifi-cation and diversification. We also provide an open-source implementa-tion of this framework which we use to model the Quadratic AssignmentProblem (QAP) with a hybrid solver, combining three metaheuristics.We present experimental evidence that PHYSH brings significant im-provements over competing approaches, as witness the performance onrepresentative hard instances of QAP |
|---|