Probabilistically distorted risk-sensitive infinite-horizon dynamic programming.
Kun LinCheng JieSteven I. MarcusPublished in: Autom. (2018)
Keyphrases
- risk sensitive
- infinite horizon
- optimal control
- dynamic programming
- markov decision processes
- finite horizon
- average cost
- markov decision problems
- optimal policy
- state space
- partially observable
- control policies
- dec pomdps
- reinforcement learning
- knapsack problem
- long run
- markov decision process
- linear program
- finite state
- policy iteration
- multistage
- linear programming
- reinforcement learning algorithms
- control strategy
- planning under uncertainty
- utility function