Reinforcement learning algorithms for semi-Markov decision processes with average reward.
Yanjie LiPublished in: ICNSC (2012)
Keyphrases
- reinforcement learning algorithms
- semi markov decision processes
- average reward
- markov decision processes
- model free
- reinforcement learning
- optimal policy
- state space
- reward function
- stochastic games
- policy iteration
- function approximation
- dynamic programming
- temporal difference
- finite state
- total reward
- policy gradient
- decision processes
- markov decision process
- partially observable
- average cost
- state variables
- markov chain
- long run
- decision problems
- learning algorithm
- infinite horizon
- reward shaping
- heuristic search