Hybridizing the 1/5-th Success Rule with Q-Learning for Controlling the Mutation Rate of an Evolutionary Algorithm.
Arina BuzdalovaCarola DoerrAnna RodionovaPublished in: CoRR (2020)
Keyphrases
- mutation rate
- evolutionary algorithm
- fitness function
- optimization problems
- multi objective
- evolutionary computation
- reinforcement learning
- learning rate
- fitness landscape
- multi agent
- cooperative
- differential evolution
- genetic programming
- multi objective optimization
- simulated annealing
- state space
- population size
- genetic algorithm
- learning algorithm
- evolutionary computing
- differential evolution algorithm
- evolution strategy
- genetic operators
- action selection
- evolutionary process
- nsga ii
- mutation operator
- crossover operator
- evolutionary search
- evolutionary strategy
- computational intelligence
- optimization method
- population dynamics
- resource allocation