报告题目：Mixtures of Gaussian copula factor analyzers for clustering high dimensional data
报 告 人： 张莉莉博士
Mixtures of factor analyzers is a useful model-based clustering method which can avoid the curse of dimensionality in high-dimensional clustering. However, this approach is sensitive to both diverse non-normalities of marginal variables and outliers, which are commonly observed in multivariate experiments. We propose mixtures of Gaussian copula factor analyzers (MGCFA) for clustering high-dimensional data.
This model has two advantages; 1) it allows dierent marginal distributions to facilitate fitting exibility of the mixture model, 2) it can avoid the curse of dimensionality by embedding the factor-analytic structure in the component-correlation matrices of the mixture distribution.
时 间：2019年10月22日（周二下午） 10：00—11:30