BriefGPT.xyz
Nov, 2021
稀疏变分高斯过程的双参数化
Dual Parameterization of Sparse Variational Gaussian Processes
HTML
PDF
Vincent Adam, Paul E. Chang, Mohammad Emtiyaz Khan, Arno Solin
TL;DR
通过使用双参数化来给每个数据示例分配双参数，以提高计算效率，使用自然梯度下降加速推理并为超参数学习提供更紧的证据下限，提高了稀疏变分高斯过程方法的计算效率。
Abstract
sparse variational gaussian process
(SVGP) methods are a common choice for non-conjugate Gaussian process inference because of their computational benefits. In this paper, we improve their
computational efficiency
→