A Reinforcement Learning Perspective on the Optimal Control of Mutation Probabilities for the (1+1) Evolutionary Algorithm: First Results on the OneMax Problem.
Luca MossinaEmmanuel RachelsonDaniel DelahayePublished in: CoRR (2019)
Keyphrases
- optimal control
- evolutionary algorithm
- reinforcement learning
- control problems
- multi objective
- evolutionary computation
- optimization problems
- differential evolution
- fitness function
- dynamic programming
- genetic programming
- feedback control
- mutation operator
- genetic algorithm
- evolutionary process
- class of nonlinear systems
- genetic operators
- control strategy
- optimal control problems
- risk sensitive
- simulated annealing
- multi objective optimization
- actor critic
- infinite horizon
- function optimization
- state space
- control law
- crossover operator
- reinforcement learning algorithms
- rl algorithms
- markov decision processes
- model free
- function approximation
- constrained optimization problems
- machine learning
- probability distribution
- optimal solution