Selector-Actor-Critic and Tuner-Actor-Critic Algorithms for Reinforcement Learning.
Ala'eddin MasadehZhengdao WangAhmed E. KamalPublished in: WCSP (2019)
Keyphrases
- actor critic
- reinforcement learning
- policy iteration
- policy gradient methods
- optimal control
- temporal difference
- reinforcement learning algorithms
- policy gradient
- approximate dynamic programming
- neuro fuzzy
- rl algorithms
- gradient method
- markov decision processes
- model free
- learning algorithm
- function approximation
- natural actor critic
- optimal policy
- average reward
- partially observable markov decision processes
- function approximators
- dynamic programming
- state space
- multi agent
- least squares
- partially observable
- convergence rate
- fixed point
- linear program
- transfer learning
- linear programming
- computational complexity
- action selection
- infinite horizon
- dynamical systems