BriefGPT.xyz
Aug, 2014
最优加权聚群是贝叶斯积分
Optimally-Weighted Herding is Bayesian Quadrature
HTML
PDF
Ferenc Huszar, David Duvenaud
TL;DR
本文研究了采样方法和贝叶斯积分估计在选取样本时的损失函数关系，进而证明了带权的 kernel herding 是所有带权采样方法中最优的方法，同时也得出了贝叶斯积分估计的经验误差上界。
Abstract
herding
and kernel
herding
are deterministic methods of choosing samples which summarise a probability distribution. A related task is choosing samples for estimating integrals using
→