BriefGPT.xyz
Apr, 2013
MCMC领域的紧缩：削减Metropolis-Hastings预算
Austerity in MCMC Land: Cutting the Metropolis-Hastings Budget
HTML
PDF
Anoop Korattikara, Yutian Chen, Max Welling
TL;DR
本研究旨在提高在处理大数据集时的贝叶斯后验MCMC采样的效率，通过引入基于连续假设检验的近似MH规则，根据部分数据即可高置信度地接受或拒绝样本，从而减少计算代价，同时控制引入的渐近偏差使方差下降得以更快。
Abstract
Can we make
bayesian posterior
mcmc sampling
more efficient when faced with very
large datasets
? We argue that computing the likelihood fo
→