Characterization of Optimal Policies in Vector-Valued Markovian Decision Processes.
Nagata FurukawaPublished in: Math. Oper. Res. (1980)
Keyphrases
- vector valued
- optimal policy
- markovian decision processes
- markov decision problems
- markov decision processes
- decision problems
- reinforcement learning
- state space
- dynamic programming
- scale space
- long run
- infinite horizon
- sufficient conditions
- average cost
- finite state
- wavelet packet
- monte carlo
- reproducing kernel hilbert space
- policy iteration
- initial state
- partially observable
- computational complexity
- markov decision process
- partially observable markov decision processes
- data mining