On discounted approximations of undiscounted stochastic games and Markov decision processes with limited randomness.
Endre BorosKhaled M. ElbassioniVladimir GurvichKazuhisa MakinoPublished in: Oper. Res. Lett. (2013)
Keyphrases
- markov decision processes
- stochastic games
- average reward
- optimal policy
- finite state
- state space
- infinite horizon
- policy iteration
- reinforcement learning algorithms
- dynamic programming
- reinforcement learning
- finite horizon
- action space
- multiagent reinforcement learning
- average cost
- markov decision process
- partially observable
- total reward
- discounted reward
- long run