On Anderson Acceleration for Partially Observable Markov Decision Processes.
Melike ErmisMingyu ParkInsoon YangPublished in: CDC (2021)
Keyphrases
- partially observable markov decision processes
- finite state
- reinforcement learning
- belief state
- dynamical systems
- optimal policy
- dynamic programming
- planning under uncertainty
- decision problems
- belief space
- continuous state
- partially observable stochastic games
- planning problems
- markov decision processes
- state space
- stochastic domains
- partial observability
- partially observable
- partially observable domains
- multi agent
- sequential decision making problems
- partially observable markov
- approximate solutions
- partially observable markov decision process
- dec pomdps
- markov chain
- infinite horizon
- initial state
- learning algorithm