Title :
Approximate reasoning with parallel connectionist production systems
Author :
Kasabov, Nikola K. ; Shishkov, Stephan I.
Author_Institution :
Dept. of Inf. Sci., Otago Univ., Dunedin, New Zealand
Abstract :
The paper presents a development of a connectionist production system NPS where a parallel updating of the working memory is introduced through parallel activation of all the fully or partially satisfied production rules. A MAX-function is used when several rules update in parallel one fact. The parallel updating in NPS is suitable for approximate reasoning over fuzzy production systems. The paper brings to one point-the point of connectionist implementation both the symbolic AI production systems and the fuzzy production systems.
Keywords :
inference mechanisms; knowledge based systems; neural nets; MAX-function; NPS; approximate reasoning; fuzzy production systems; parallel connectionist production systems; symbolic AI production systems; working memory parallel updating; Artificial intelligence; Computer science; Fires; Fuzzy logic; Fuzzy reasoning; Fuzzy systems; Information science; Neural networks; Parallel processing; Production systems;
Conference_Titel :
Neural Networks, 1993. IJCNN '93-Nagoya. Proceedings of 1993 International Joint Conference on
Print_ISBN :
0-7803-1421-2
DOI :
10.1109/IJCNN.1993.714344