Discussion of “Influential features PCA for high dimensional clustering”
نویسندگان
چکیده
منابع مشابه
Discussion of Influential Features Pca for High Dimensional Clustering
We commend Jin and Wang on a very interesting paper introducing a novel approach to feature selection within clustering and a detailed analysis of its clustering performance under a Gaussian mixture model. I shall divide my discussion into several parts: (i) prior work on feature selection and clustering; (ii) theoretical aspects; (iii) practical aspects; and finally (iv) some questions and dir...
متن کاملInfluential Features Pca for High Dimensional Clustering
We consider a clustering problem where we observe feature vectors Xi ∈ R, i = 1, 2, . . . , n, from K possible classes. The class labels are unknown and the main interest is to estimate them. We are primarily interested in the modern regime of p n, where classical clustering methods face challenges. We propose Influential Features PCA (IF-PCA) as a new clustering procedure. In IF-PCA, we select...
متن کاملDiscussion of “ Influential Feature Pca for High Dimensional Clustering ”
We would like to congratulate the authors for an interesting paper and a novel proposal for clustering high-dimensional Gaussian mixtures with a diagonal covariance matrix. The proposed two-stage procedure first selects features based on the Kolmogorov-Smirnov statistics and then applies a spectral clustering method to the post-selected data. A rigorous theoretical analysis for the clustering e...
متن کاملDiscussion of “ Influential Features Pca for High Dimensional Clustering ” , by J . Jin And
where z : {1, . . . , n} → {1, . . . ,K} is an unknown assignment of the observations to K classes, μ1, . . . , μK are unknown vectors in Rp, and Zi ∈ Rp are i.i.d. normal vectors with mean 0 and covariance matrix σIp. Here Ip is the p× p identity matrix. In [JW], the covariance matrix of Z1 is diagonal, with the diagonal elements bounded from below and from above by constants red that are inde...
متن کاملImportant Features PCA for high dimensional clustering
We consider a clustering problem where we observe feature vectors Xi ∈ R, i = 1, 2, . . . , n, from K possible classes. The class labels are unknown and the main interest is to estimate them. We are primarily interested in the modern regime of p n, where classical clustering methods face challenges. We propose Important Features PCA (IF-PCA) as a new clustering procedure. In IFPCA, we select a ...
متن کاملذخیره در منابع من
با ذخیره ی این منبع در منابع من، دسترسی به آن را برای استفاده های بعدی آسان تر کنید
ژورنال
عنوان ژورنال: The Annals of Statistics
سال: 2016
ISSN: 0090-5364
DOI: 10.1214/16-aos1423a