Deep Random Splines for Point Process Intensity Estimation of Neural Population Data

DRS are a deep probabilistic model in which standard Gaussian noise is transformed through a neural network to obtain the parameters of a spline, and the random function is then the corresponding spline.