Global Linear Convergence of Online Reinforcement Learning for Partially Observable Systems.
Takumi HiraiTomonori SadamotoPublished in: ECC (2022)
Keyphrases
- partially observable
- reinforcement learning
- state space
- partial observability
- markov decision processes
- hidden state
- partially observable domains
- partially observable environments
- markov decision problems
- dynamical systems
- belief state
- complex systems
- infinite horizon
- decision problems
- action models
- initially unknown
- fully observable
- partial observations
- optimal policy
- np hard