搜索结果: 1-5 共查到“管理学 Mixture model”相关记录5条 . 查询时间(0.062 秒)
A non-parametric mixture model for topic modeling over time
non-parametric mixture model topic modeling over time
font style='font-size:12px;'>
2012/9/17
A single, stationary topic model such as la-tent Dirichlet allocation is inappropriate for modeling corpora that span long time peri-ods, as the popularity of topics is likely to change over time. A n...
Splitting and Merging Components of a Nonconjugate Dirichlet Process Mixture Model
Bayesian model Markov chain Monte Carlo split-merge moves nonconjugate prior
font style='font-size:12px;'>
2009/9/22
The inferential problem of associating data to mixture components is dif-
ficult when components are nearby or overlapping. We introduce a new split-merge
Markov chain Monte Carlo technique that eff...
Convergence properties of a general algorithm for calculating variational Bayesian estimates for a normal mixture model
Mixture model Variational Bayes Local convergence Laplace approximation
font style='font-size:12px;'>
2009/9/21
In this paper we propose a generalised iterative algorithm for calcu-
lating variational Bayesian estimates for a normal mixture model and investigate
its convergence properties. It is shown theoret...
Mean field inference for the Dirichlet process mixture model
Bayesian nonparametrics approximation methods variational inference density estimation
font style='font-size:12px;'>
2009/9/16
We present a systematic study of several recently proposed methods of mean field inference for the Dirichlet process mixture (DPM) model. These methods provide approximations to the posterior distribu...
A mixture model for unsupervised tail estimation
Mixing functions Heavy and Light Tailed Distributions Maximum Likelihood Mixture Mod
font style='font-size:12px;'>
2010/3/18
This paper proposes a new method to combine several densities
such that each density dominates a separate part of a joint distribution. The
method is fully unsupervised, i.e. the parameters in the d...