Model-Based Reinforcement Learning Exploiting State-Action Equivalence.
Mahsa AsadiMohammad Sadegh TalebiHippolyte BourelOdalric-Ambrym MaillardPublished in: CoRR (2019)
Keyphrases
- model based reinforcement learning
- state action
- markov decision processes
- reinforcement learning
- average reward
- action space
- stochastic games
- markov decision process
- evaluation function
- optimal policy
- reward function
- state space
- policy iteration
- finite state
- function approximation
- dynamic programming
- multi agent
- function approximators
- neural network
- state transitions
- partially observable
- temporal difference
- infinite horizon
- optimal control
- markov random field