Learning to Act in Decentralized Partially Observable MDPs.
Jilles Steeve DibangoyeOlivier BuffetPublished in: ICML (2018)
Keyphrases
- partially observable
- reinforcement learning
- markov decision processes
- action models
- state space
- markov decision problems
- hidden state
- dynamical systems
- decision problems
- learning algorithm
- partially observable environments
- partial observations
- infinite horizon
- partial observability
- belief state
- fully observable
- partially observable domains
- dec pomdps
- average reward
- decision theoretic
- random variables
- special case