Optimizing Memory-Bounded Controllers for Decentralized POMDPs.
Christopher AmatoDaniel S. BernsteinShlomo ZilbersteinPublished in: UAI (2007)
Keyphrases
- reinforcement learning
- distributed constraint optimization
- dec pomdps
- multi agent
- cooperative
- partially observable markov decision processes
- control system
- memory requirements
- computing power
- bounded memory
- continuous state
- memory usage
- peer to peer
- limited memory
- infinite horizon
- control law
- partially observable
- optimal control
- memory management
- markov decision processes
- dynamical systems
- search algorithm