LimNet-Flexible Learning Network Containing Linear Properties
Jinglu Hu*, Kotaro Hirasawa* and Kousuke Kumamaru**
*Department of Electrical and Electronic Systems Engineering, Kyushu University 6-10-1 Hakozaki, Higashi-ku, Fukuoka 812-8581, Japan
**Department of Control Engineering and Science, Kyushu Institute of Technology 680-4 Kawatsu, lizuka 812, Japan
Received:September 25, 1998Accepted:March 15, 1999Published:August 20, 1999
Keywords:Learning network, Linear model, Prior knowledge, Modeling and identification, Nonlinear system
We propose a flexible learning network of a class of linear models. A nonlinear black box system is transformed into a network of known and unknown nodes (node functions), where a linear model is introduced. Unknown nodes are parameterized using neurofuzzy models. The resulting learning network is interpreted as a linear model network (LimNet) and features useful linear properties and universal approximation ability.
Cite this article as:J. Hu, K. Hirasawa, and K. Kumamaru, “LimNet-Flexible Learning Network Containing Linear Properties,” J. Adv. Comput. Intell. Intell. Inform., Vol.3 No.4, pp. 303-311, 1999.Data files: