抄録
This paper proposes a new computation method of the renewal function applying a radial basis function type of neural network. The basic idea is to approximate the Stieltjes convolution of a distribution function by the neural network, based on the similar concept to the Spline theory. Throughout numerical experiments, we show that the proposed method can numerically calculate the renewal function with higher precision than existing methods such as the cubic Spline algorithm and the discretization algorithm.
