BriefGPT.xyz
Jun, 2021
基于谐核分解的可扩展变分高斯过程
Scalable Variational Gaussian Processes via Harmonic Kernel Decomposition
HTML
PDF
Shengyang Sun, Jiaxin Shi, Andrew Gordon Wilson, Roger Grosse
TL;DR
本文提出了一种新的可扩展的变分高斯过程近似方法，采用傅里叶级数将核作为正交核的和进行分解，利用这种正交性使众多诱导点具有低计算成本，能在回归和分类问题中利用输入空间的对称性，明显优于标准变分方法，并在CIFAR-10中取得与纯GP模型的最新成果。
Abstract
We introduce a new
scalable variational gaussian process approximation
which provides a high fidelity approximation while retaining general applicability. We propose the
harmonic kernel decomposition
(HKD), which
→