An Average-Reward Reinforcement Learning Algorithm for Computing Bias-Optimal Policies.
Sridhar MahadevanPublished in: AAAI/IAAI, Vol. 1 (1996)
Keyphrases
- average reward
- optimal policy
- markov decision processes
- long run
- average reward reinforcement learning
- semi markov decision processes
- reinforcement learning
- discounted reward
- state space
- policy iteration
- decision problems
- dynamic programming
- total reward
- optimality criterion
- multistage
- sufficient conditions
- finite horizon
- model free
- sample path
- infinite horizon
- markov chain
- finite state
- markov decision process
- partially observable markov decision processes
- state and action spaces
- average cost
- hierarchical reinforcement learning
- discount factor
- machine learning