The Biased Random-Key Genetic Algorithm (BRKGA) is an efficient metaheuristic to solve combinatorial optimization problems but requires parameter tuning so the intensification and diversification of the algorithm work in a balanced way. There is, however, not only one optimal parameter configuration, and the best configuration may differ according to the stages of the evolutionary process. Hence, in this research paper, a BRKGA with Q-Learning algorithm (BRKGA-QL) is proposed. The aim is to control the algorithm parameters during the evolutionary process using Reinforcement Learning, indicating the best configuration at each stage. In the experiments, BRKGA-QL was applied to the symmetric Traveling Salesman Problem, and the results show the efficiency and competitiveness of the proposed algorithm.
Citation
Chaves, A.A. and Lorena, L.H.N. An adaptive and near parameter-free BRKGA using Reinforcement Learning. Technical Report. UNIFESP, 2021.