ASYMPTOTIC NORMALITY OF MAXIMUM LIKELIHOOD AND ITS VARIATIONAL APPROXIMATION FOR STOCHASTIC BLOCKMODELS
ASYMPTOTIC NORMALITY OF MAXIMUM LIKELIHOOD AND ITS VARIATIONAL APPROXIMATION FOR STOCHASTIC BLOCKMODELS
复制标题
随机块模型的极大似然渐近正态性及其变分逼近
DOI：
10.1214/13-aos1124
复制
发表时间：
2013-08-01
影响因子：
4.5
通讯作者：
Zhang, Hai
中科院分区：
文献类型：
--
作者：
Bickel, Peter;Choi, David;Zhang, Hai
参数估计的变分方法是一个活跃的研究领域，可能提供计算上易于处理的算法与理论性能界限。我们建立在最近的工作，将这种方法应用于网络数据，并建立随机块模型数据的参数估计的渐近正态率，无论是最大似然或变分估计。该结果也适用于文献中发现的随机块模型的各种子模型。
Variational methods for parameter estimation are an active research area, potentially offering computationally tractable heuristics with theoretical performance bounds. We build on recent work that applies such methods to network data, and establish asymptotic normality rates for parameter estimates of stochastic blockmodel data, by either maximum likelihood or variational estimation. The result also applies to various sub-models of the stochastic blockmodel found in the literature.