Recursive Learning Automata for Control of Partially Observable Markov Decision Processes.
Hyeong Soo ChangMichael C. FuSteven I. MarcusPublished in: CDC/ECC (2005)
Keyphrases
- learning automata
- partially observable markov decision processes
- reinforcement learning
- pursuit algorithm
- learning automaton
- finite state
- continuous state
- optimal policy
- stochastic domains
- dynamical systems
- belief state
- state space
- multi agent
- control system
- planning under uncertainty
- partially observable stochastic games
- decision problems
- dynamic programming
- belief space
- probability distribution
- control policies
- infinite horizon
- machine learning
- control strategy
- planning problems
- dynamic environments