SUMMATION-INTEGRAL BERNSTEIN TYPE OF NEURAL NETWORK OPERATORS

In this paper, we introduce a family of neural network operators of summation-integral Bernstein type, which are define by using some sigmoidal functions. We give pointwise and uniform approximation theorems for these operators when are applied for continuous functions. In addition, we discuss the approximation for these operators in -spaces with . Then we give some applications of the sequences of a family of linear positive multivariate neural network operators , then we describe the results by graphics of the error function for some particular values of and for two test functions in -dimensional.