When Is Partially Observable Reinforcement Learning Not Scary?
Qinghua LiuAlan ChungCsaba SzepesváriChi JinPublished in: CoRR (2022)
Keyphrases
- partially observable
- reinforcement learning
- state space
- markov decision processes
- partial observability
- partially observable domains
- partially observable environments
- markov decision problems
- partial observations
- dynamical systems
- function approximation
- action models
- model free
- hidden state
- infinite horizon
- decision problems
- reward function
- reinforcement learning algorithms
- belief space
- multi agent
- temporal difference
- optimal policy
- learning algorithm
- policy iteration
- partially observable markov decision process
- machine learning
- markov chain
- control system
- orders of magnitude
- computational complexity
- partially observable markov decision processes
- heuristic search