이질성 구조를 갖는 신경회로망에 관한 연구 = A Study of Neural Networks with Heterogeneous Structures

다국어 초록 (Multilingual Abstract)

There are two methods to imitate a human's brain. One is a symbolic processing artifical intelligence which programs knowledges in brains according to functions. The other is the artificial intelligence intending to implement the intelligence by using the brain's organization achieved from brain analysis. Between these, a symbolic processing artificial intelligence is the approach of level of functions-called Fuzzy Logic in other word. And we say it Neural Network the approach of the level of physiology.
In this paper, a new fuzzy learning method of improved learing speed and convergence rate is proposed to exploit the advantages of artificial neural networks and fuzzy logic to neuron structure. This method is applied to the exclusive OR problem, Parity problem which is used as the benchmark in neurla network. As a result of the experiment, it does ot always guarantee the convergence. However, the proposed network shows an improved learning time and a high convergence rate. The proposed network is albe to extend the arbitrary layer. When we consider only a network of the single layer, the network has the capability of high speed during learning process and rapid processin on huge patterns.