Speeding Up the Convergence of Value Iteration in Partially Observable Markov Decision Processes.
Nevin Lianwen ZhangWeihong ZhangPublished in: J. Artif. Intell. Res. (2001)
Keyphrases
- partially observable markov decision processes
- finite state
- partially observable markov
- reinforcement learning
- belief space
- belief state
- dynamical systems
- planning under uncertainty
- markov decision processes
- decision problems
- optimal policy
- dynamic programming
- continuous state
- state space
- planning problems
- multi agent
- partially observable stochastic games
- convergence rate
- stochastic domains
- sequential decision making problems
- stochastic shortest path
- infinite horizon
- partially observable domains
- average reward
- partially observable
- initial state
- approximate solutions
- model checking
- heuristic search
- markov decision process
- computational complexity
- long run
- belief revision
- orders of magnitude
- learning algorithm