BriefGPT.xyz
Nov, 2012
一类相关随机测度的统一表示
A unifying representation for a class of dependent random measures
HTML
PDF
Nicholas J. Foti, Joseph D. Futoma, Daniel N. Rockmore, Sinead Williamson
TL;DR
该研究依据拓展空间中的稀疏泊松过程，提出了依赖随机测度的一般建模方法，并应用到完全随机测度模型。该框架可应用于多种模型中，且构建的具有协变性的潜在特征模型和主题模型表现出更好的预测性能。
Abstract
We present a general construction for
dependent random measures
based on thinning
poisson processes
on an augmented space. The framework is not restricted to dependent versions of a specific nonparametric model,
→