Solution to the risk-sensitive average cost optimality equation in a class of Markov decision processes with finite state space.
Rolando Cavazos-CadenaPublished in: Math. Methods Oper. Res. (2003)
Keyphrases
- average cost
- finite state
- markov decision processes
- risk sensitive
- markov decision chains
- approximate dynamic programming
- optimal policy
- control policy
- action space
- markov chain
- finite horizon
- state space
- reinforcement learning
- finite number
- initial state
- optimal control
- policy iteration
- dynamic programming
- long run
- mathematical model
- action sets
- planning under uncertainty
- reinforcement learning algorithms
- partially observable markov decision processes
- infinite horizon
- markov decision process
- decision processes
- optimal solution
- linear programming
- partially observable
- reward function
- control policies
- model checking
- total cost
- stationary policies
- least squares
- search space