Evaluating Correctness of Reinforcement Learning based on Actor-Critic Algorithm.
Youngjae KimManzoor HussainJae-Won SuhJang-Eui HongPublished in: ICUFN (2022)
Keyphrases
- reinforcement learning
- actor critic
- computational complexity
- dynamic programming
- learning algorithm
- cost function
- objective function
- simulated annealing
- policy gradient
- machine learning
- search space
- optimal solution
- convergence rate
- model free
- np hard
- linear programming
- particle swarm optimization
- policy iteration
- control policy
- convergence proof