FUNCTION APPROXIMATION USING FUZZY NEURAL NETWORKS WITH ROBUST LEARNING ALGORITHM

The paper describes a novel application of the B-spline membership fun ctions (BMF's) and the fuzzy neural network to the function approximat ion with outliers in training data, According to the robust objective function, we use gradient descent method to derive the new learning ru les of the weighting values and BMF's of the fuzzy neural network for robust function approximation, In this paper, the robust learning algo rithm is derived, During the learning process, the robust objective fu nction comes into effect and the approximated function will gradually be unaffected by the erroneous training data, As a result, the robust function approximation can rapidly converge to the desired tolerable e rror scope, In other words, the learning iterations will decrease grea tly, We realize the function approximation not only in one dimension ( curves), but also in two dimension (surfaces), Several examples are si mulated in order to confirm the efficiency and feasibility of the prop osed approach in this paper.