Weighted reward criteria in Competitive Markov Decision Processes.
Jerzy A. FilarO. J. VriezePublished in: ZOR Methods Model. Oper. Res. (1992)
Keyphrases
- markov decision processes
- average reward
- reinforcement learning
- reward function
- optimality criterion
- discounted reward
- total reward
- expected reward
- optimal policy
- finite state
- state space
- policy iteration
- transition matrices
- reinforcement learning algorithms
- dynamic programming
- stationary policies
- reachability analysis
- planning under uncertainty
- semi markov decision processes
- average cost
- infinite horizon
- decision processes
- finite horizon
- function approximation
- stochastic games
- decision diagrams
- risk sensitive
- long run
- partially observable
- markov decision process
- model based reinforcement learning
- action sets
- learning algorithm
- state and action spaces
- decision theoretic planning
- hierarchical reinforcement learning
- optimal control
- decision problems
- policy evaluation
- decision making