Neuro-Fuzzy Function Approximations Using Feedforward Networks - An Application of Sigmoidal Signal

摘要

Chen, Cybenko, Funahashi, Babri, Gorban, Barron and others studied function approximation capabilities by feed forward neural networks. Ramakrishnan and his collaborators introduced the left sigmoidal and right sigmoidal signals to prove some function approximation theorems in continuous functions. In this paper, the right sigmoidal signal is used to establish function approximation theorems using infimum and supremum operators.