Hybridizing the 1/5-th Success Rule with Q-Learning for Controlling the Mutation Rate of an Evolutionary Algorithm.
Arina BuzdalovaCarola DoerrAnna RodionovaPublished in: PPSN (2) (2020)
Keyphrases
- mutation rate
- evolutionary algorithm
- fitness function
- multi objective
- evolutionary computation
- optimization problems
- reinforcement learning
- differential evolution
- learning rate
- simulated annealing
- fitness landscape
- population size
- genetic algorithm
- genetic programming
- learning algorithm
- multi agent
- mutation operator
- state space
- multi objective optimization
- cooperative
- evolutionary computing
- population dynamics
- optimization method
- action selection
- genetic operators
- evolutionary search
- evolutionary strategy
- evolutionary process
- crossover operator
- differential evolution algorithm
- particle swarm optimization pso
- neural network
- particle swarm
- multi agent reinforcement learning
- nsga ii
- parameter settings
- convergence rate
- dynamic programming
- computational intelligence