NeuralEF: Deconstructing Kernels by Deep Neural Networks. (arXiv:2205.00165v3 [cs.LG] UPDATED)

Web: http://arxiv.org/abs/2205.00165

Learning the principal eigenfunctions of an integral operator defined by a
kernel and a data distribution is at the core of many machine learning
problems. Traditional nonparametric solutions based on the Nystr{\"o}m formula
suffer from scalability issues. Recent work has resorted to a parametric
approach, i.e., training neural networks to approximate the eigenfunctions.
However, the existing method relies on an expensive orthogonalization step and
is difficult to implement. We show that these problems can be fixed by using a
new …

arxiv deep lg networks neural neural networks