BriefGPT.xyz
Sep, 2017
卷积高斯过程
Convolutional Gaussian Processes
HTML
PDF
Mark van der Wilk, Carl Edward Rasmussen, James Hensman
TL;DR
介绍了在 Gaussian 过程中引入卷积结构的方法，并构建了适用于卷积核的跨域诱导点逼近，利用较快但准确的后验推理获得卷积核的泛化利益，应用于多项研究中，其中通过边缘似然可进一步提高性能。
Abstract
We present a practical way of introducing
convolutional structure
into
gaussian processes
, making them more suited to high-dimensional inputs like images. The main contribution of our work is the construction of
→