Title :
Extended theory refinement in knowledge-based neural networks
Author :
Garcez, Artur S d Avila
Author_Institution :
Dept. of Comput., Imperial Coll. of Sci., Technol. & Med., London, UK
fDate :
6/24/1905 12:00:00 AM
Abstract :
This paper shows that single hidden layer networks with semi-linear activation function compute the answer set semantics of extended logic programs. As a result, incomplete (nonmonotonic) theories, presented as extended logic programs, i.e., possibly containing both classical and default negations, may be refined through inductive learning in knowledge-based neural networks
Keywords :
feedforward neural nets; knowledge based systems; learning (artificial intelligence); logic programming; transfer functions; default negation; extended logic programming; feedforward neural networks; hybrid architectures; inductive learning; knowledge-based neural networks; semilinear activation function; single hidden layer networks; translation algorithm; Artificial neural networks; Computer networks; Concurrent computing; Educational institutions; Hybrid power systems; Intelligent networks; Logic programming; Machine learning; Neural networks; Refining;
Conference_Titel :
Neural Networks, 2002. IJCNN '02. Proceedings of the 2002 International Joint Conference on
Conference_Location :
Honolulu, HI
Print_ISBN :
0-7803-7278-6
DOI :
10.1109/IJCNN.2002.1007610
