A brief note on understanding neural networks as Gaussian processes
Machine Learning
2021-07-27 v1 Computational Engineering, Finance, and Science
Machine Learning
Abstract
As a generalization of the work in [Lee et al., 2017], this note briefly discusses when the prior of a neural network output follows a Gaussian process, and how a neural-network-induced Gaussian process is formulated. The posterior mean functions of such a Gaussian process regression lie in the reproducing kernel Hilbert space defined by the neural-network-induced kernel. In the case of two-layer neural networks, the induced Gaussian processes provide an interpretation of the reproducing kernel Hilbert spaces whose union forms a Barron space.
Cite
@article{arxiv.2107.11892,
title = {A brief note on understanding neural networks as Gaussian processes},
author = {Mengwu Guo},
journal= {arXiv preprint arXiv:2107.11892},
year = {2021}
}