Unifying Nondeterministic and Probabilistic Planning Through Imprecise Markov Decision Processes.
Felipe W. TrevizanFábio Gagliardi CozmanLeliane Nunes de BarrosPublished in: IBERAMIA-SBIA (2006)
Keyphrases
- probabilistic planning
- markov decision processes
- finite state
- planning under uncertainty
- decision theoretic planning
- state space
- optimal policy
- markov decision process
- partially observable
- policy iteration
- dynamic programming
- reinforcement learning
- average cost
- infinite horizon
- average reward
- initial state
- decision processes
- heuristic search
- finite automata
- reinforcement learning algorithms
- action space
- multistage
- partially observable markov decision processes
- reward function