Perturbed Generalized Logistic Activated Singular Integrals Multivariate Approximation with Rates
摘要
Here we treat the quantitative multivariate approximation of perturbed generalized logistic activated singular integral operators to the unit operator. The engaged neural network activation function is both parametrized and deformed and the related kernel is a density function on \(\mathbb {R}^{N}\) . We exhibit uniform and \(L_{p}\) , \(p\ge 1\) , approximations via Jackson type inequalities involving the first \(L_{p}\) modulus of smoothness, \(1\le p\le \infty \) . Differentiability of our multivariate functions is covered extensively in our approximations.