Mixed risk-neutral/minimax control of discrete-time, finite-state Markov decision processes.
Stefano P. CoraluppiSteven I. MarcusPublished in: IEEE Trans. Autom. Control. (2000)
Keyphrases
- finite state
- markov decision processes
- risk sensitive
- risk neutral
- optimal policy
- state space
- reinforcement learning
- dynamic programming
- markov chain
- policy iteration
- reinforcement learning algorithms
- decision processes
- average cost
- risk averse
- infinite horizon
- action sets
- partially observable
- finite horizon
- model checking
- control system
- markov decision process
- average reward
- policy iteration algorithm
- partially observable markov decision processes
- learning algorithm
- optimal control
- control strategy
- decision making
- machine learning