Generate matrix-wise functions for two matrices using neural network kernel.
kernel_nn(l, p, sigma)(numeric) A numeric number indicating the hyperparameter (flexibility) of a specific kernel.
(integer) For polynomial, p is the power; for matern, v = p + 1 / 2; for rational, alpha = p.
(numeric) The covariance coefficient for neural network kernel.
(function) A function calculating the relevance of two matrices.
Neural Network Kernels $$k_{NN}(x, x')=\frac{2}{\pi}sin^{-1}\Big(\frac{2\tilde{x}^T \tilde{x}'}{\sqrt{(1+2\tilde{x}^T \tilde{x})(1+2\tilde{x}'^T \tilde{x}')}}\Big)$$
The MIT Press. Gaussian Processes for Machine Learning, 2006.