论文部分内容阅读
Consider a two-class clustering problem where we have the data Xi = liμ+Zi,Zi~N(0,Ip),1 ≤ i ≤ n.The class labels li ∈ {-1,1} are unknown and the main interest is to estimate them.The feature vector μ ∈ Rp is also unknown but is presumably sparse in that only a small fraction of the features are useful for clustering.