Solutions of the average cost optimality equation for finite Markov decision chains: risk-sensitive and risk-neutral criteria.
Rolando Cavazos-CadenaPublished in: Math. Methods Oper. Res. (2009)
Keyphrases
- risk sensitive
- markov decision chains
- risk neutral
- average cost
- optimality criterion
- finite number
- markov decision processes
- optimal control
- long run
- finite state
- utility function
- optimal policy
- infinite horizon
- finite horizon
- linear program
- evaluation function
- optimal solution
- average reward
- model free
- state space
- reinforcement learning
- initial state
- dynamic programming
- total cost
- linear programming
- control policy
- markov decision problems
- expected utility
- markov decision process
- policy iteration
- machine learning
- risk averse
- np hard
- risk aversion
- lower bound
- multistage