Risk-sensitive finite-horizon piecewise deterministic Markov decision processes.
Yonghui HuangZhaotong LianXianping GuoPublished in: Oper. Res. Lett. (2020)
Keyphrases
- markov decision processes
- risk sensitive
- finite horizon
- optimal policy
- average cost
- infinite horizon
- finite state
- state space
- dynamic programming
- control policies
- policy iteration
- markov decision process
- partially observable
- reinforcement learning
- decision processes
- reward function
- planning under uncertainty
- reinforcement learning algorithms
- decision problems
- average reward
- bayesian networks
- markov decision problems
- initial state
- data mining
- decision making
- learning algorithm
- machine learning