Reward Revision for Discounted Markov Decision Problems.
Chelsea C. White IIILyn C. ThomasWilliam T. SchererPublished in: Oper. Res. (1985)
Keyphrases
- markov decision problems
- average reward
- optimal policy
- markov decision processes
- reinforcement learning
- long run
- infinite horizon
- reward function
- policy iteration
- average cost
- partially observable
- state space
- dynamic programming
- reward shaping
- markov decision process
- finite horizon
- belief revision
- decision problems
- model free
- finite state
- decision processes
- transition probabilities
- linear programming
- learning agent
- multistage
- reinforcement learning algorithms
- action space
- initial state
- markov chain
- optimal control
- knowledge base
- function approximation
- decision theoretic
- temporal difference
- least squares
- decision making
- machine learning
- computational complexity
- learning algorithm