DocumentCode :
2358229
Title :
Q-concept-learning: generalization with concept lattice representation in reinforcement learning
Author :
Ricordeau, Marc
Author_Institution :
Laboraoire d´´Informatique, de Robotique ed de Microelectronique de Montpellier, France
fYear :
2003
fDate :
3-5 Nov. 2003
Firstpage :
316
Lastpage :
323
Abstract :
One of the very interesting properties of reinforcement learning algorithms is that they allow learning without prior knowledge of the environment. However, when the agents use algorithms that enable a generalization of the learning, they are unable to explain their choices. Neural networks are good examples of this problem. After a reminder about the basis of reinforcement learning, the lattice concept will be introduced. Then, Q-concept-learning, a reinforcement learning algorithm that enables a generalization of the learning, the use of structured languages as well as an explanation of the agent´s choices will be presented.
Keywords :
knowledge representation; learning (artificial intelligence); neural nets; Q-concept-learning; concept lattice representation; learning generalization; neural network; reinforcement learning; structured language; Artificial intelligence; Convergence; Delay; Equations; Lattices; Learning; Neural networks; Robots; Testing;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
Tools with Artificial Intelligence, 2003. Proceedings. 15th IEEE International Conference on
ISSN :
1082-3409
Print_ISBN :
0-7695-2038-3
Type :
conf
DOI :
10.1109/TAI.2003.1250206
Filename :
1250206
Link To Document :
بازگشت