Actor-Critic for Linearly-Solvable Continuous MDP with Partially Known Dynamics.
Tomoki NishiPrashant DoshiMichael R. JamesDanil V. ProkhorovPublished in: CoRR (2017)
Keyphrases
- actor critic
- reinforcement learning
- policy iteration
- markov decision processes
- average reward
- special case
- policy gradient
- optimal control
- approximate dynamic programming
- optimal policy
- reinforcement learning algorithms
- action space
- np hard
- dynamical systems
- temporal difference
- markov decision process
- function approximation
- partially observable markov decision processes
- gradient method
- neuro fuzzy
- state space
- linear program
- partially observable
- finite state
- linear programming
- dynamic programming
- computational complexity
- fixed point
- fuzzy logic
- rl algorithms
- optimal solution
- multi agent