Title :
A hybrid neural network using fuzzy reference points as hidden units
Author :
Huang, Jingwei ; Wang, Zhongtuo
Author_Institution :
Sch. of Econ. & Manage., Tsinghua Univ., Beijing, China
Abstract :
In this paper, a hybrid multilayer feedforward neural network in which hidden units are defined as fuzzy reference points, and a general feedback learning method are presented. This model has been used in a DSS. Results of an experiment on process identification are given. The proposed model is able to overcome some limitations of conventional multilayer feedforward neural networks applied in expert systems.
Keywords :
feedback; feedforward neural nets; fuzzy logic; identification; learning (artificial intelligence); multilayer perceptrons; DSS; expert systems; fuzzy reference points; general feedback learning method; hidden units; hybrid multilayer feedforward neural network; process identification; Decision support systems; Expert systems; Feedforward neural networks; Fuzzy neural networks; Fuzzy sets; Multi-layer neural network; Neural networks; Neurofeedback; Power generation economics; Systems engineering and theory;
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.1993.714354
