Efficient actor-critic algorithm with dual piecewise model learning.
Shan ZhongQuan LiuShengrong GongQi-ming FuJin XuPublished in: SSCI (2017)
Keyphrases
- actor critic
- learning algorithm
- probabilistic model
- mathematical model
- gradient method
- cost function
- objective function
- reinforcement learning
- dynamic programming
- kalman filter
- model free
- policy gradient
- optimal control
- supervised learning
- simulated annealing
- monte carlo
- learning tasks
- step size
- dynamic bayesian networks
- computational complexity
- approximate dynamic programming
- optimal solution