Optimal control of average reward constrained continuous-time finite Markov decision processes.
Eugene A. FeinbergPublished in: CDC (2002)
Keyphrases
- optimal control
- average reward
- markov decision processes
- dynamic programming
- state and action spaces
- reinforcement learning
- policy iteration
- optimal policy
- control problems
- state space
- discounted reward
- semi markov decision processes
- stochastic games
- optimality criterion
- finite state
- infinite horizon
- optimal control problems
- action space
- risk sensitive
- control strategy
- state action
- decision theoretic planning
- total reward
- model free
- finite horizon
- stationary policies
- markov decision process
- hierarchical reinforcement learning
- function approximation
- linear programming
- reinforcement learning algorithms
- long run
- policy iteration algorithm
- average cost
- policy gradient
- decision processes
- temporal difference
- finite number
- markov chain
- learning algorithm
- data mining