An Adaptive Architecture for Modular Q-Learning.

Takayuki Kohri Kei Matsubayashi Mario Tokoro

Published in: IJCAI (2) (1997)

Keyphrases

reinforcement learning
cooperative
real time
management system
modular architecture
optimal policy
network architecture
highly flexible
dynamic programming
software architecture
action selection
reinforcement learning algorithms
loosely coupled
modular neural networks