APPLICATION OF FUZZY NEURAL COMPUTING IN CIRCUIT PARTITIONING
Cad La · Chinese Journal of Computers · 1992
Fuzzy set theory is adopted in a new neural computing algorithm-fuzzy neural' computing. It has been proved that the new algorithm gets rid of two fatal defects facing most of the existing neural computing algorithms, known as the slow training speed and network parameter sensibility. It is well suited to general pattern classification tasks. Application to the partitioning of electrical circuits show that Fuzzy neural computing is superior to the Kohonen's self-organizing neural computing both in the sense of training speed and final results.