A method to implement qualitative knowledge in multi-layered neural network
We propose an improved synthesis method for the multi-layered neural network(NN) using a ”translation mechanism” that maps the logical representation of qualitative knowledge into a multi-layered NN structure. We give realizability conditions and synthesis equations to realize logical functions by the NN.
The NN is tuned by back-propagation(BP) after the direct synthesis. This direct synthesis decreases the burden on the BP and contributes to the efficiency and accuracy of BP learning process.
We demonstrate our method through function approximation and character recognition problems.
Unable to display preview. Download preview PDF.
- 1.D.E.Rumelhart, J.L.McClelland and the PDP Research Group Eds., Parallel Distributed Processing, MA:MIT Press, 1986Google Scholar
- 2.B.Kosko, Neural Networks And Fuzzy Systems, NJ: Prentice Hall, 1992Google Scholar
- 3.R.Serra and G.Zanarini, Complex Systems and Cognitive Processes, Heidelberg:Springer-Verlag, 1990Google Scholar
- 4.Special Issue on Neural Network, Journal of the Society of Instrument and Control Engineers, Vol.30, No.4, 1991 (in Japanese)Google Scholar
- 6.I.Hayashi, H.Nomura, N.Wakami, ”Artificial neural network driven fuzzy control and its application to the learning of inverted pendulum system”, presented at UFSA '89 Seattle, 610–613, 1989Google Scholar
- 7.M.M. Gupta, and M.B. Gorzalczny, ”Fuzzy neuro-computational techniques and its application to modelling and control”, Presented at IFSA'91, Brussels, Proceeding Vol. Artificail Intelligence, 46–49, 1991Google Scholar
- 8.R.R. Yager, ”Using fuzzy logic to build neural networks”, Presented at IFSA'91, Brussels, Proceeding Vol. Artificail Intelligence, 210–213, 1991Google Scholar
- 9.H. Ishibuchi, R.Fujioka, H.Tanaka, ”Possibility and necessity data analysis using neural networks”, Presented at IFSA'91, Brussels, Proceeding Vol. Artificail Intelligence, 74–77, 1991Google Scholar
- 10.R.Fujioka, H. Ishibuchi, H.Tanaka, M. Omae, ”Learning algorithm of neural networks for interval-valued data”, Presented at IFSA'91, Brussels, Proceeding Vol. Artificail Intelligence, 37–40, 1991Google Scholar