Support Recovery in Sparse PCA with Incomplete Data

• URL: http://arxiv.org/abs/2205.15215v1
• Date: Mon, 30 May 2022 16:17:39 GMT
• Title: Support Recovery in Sparse PCA with Incomplete Data
• Authors: Hanbyul Lee, Qifan Song, Jean Honorio
• Abstract summary: We use a practical algorithm for principal component analysis (PCA) incomplete data. We provide theoretical and experimental evidence that the unknown true SDP enables exactly recover an incomplete support matrix. We validate our theoretical results with incomplete data, and show encouraging meaningful results on a variance expression.
• Score: 27.3669650952144
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: We study a practical algorithm for sparse principal component analysis (PCA) of incomplete and noisy data. Our algorithm is based on the semidefinite program (SDP) relaxation of the non-convex $l_1$-regularized PCA problem. We provide theoretical and experimental evidence that SDP enables us to exactly recover the true support of the sparse leading eigenvector of the unknown true matrix, despite only observing an incomplete (missing uniformly at random) and noisy version of it. We derive sufficient conditions for exact recovery, which involve matrix incoherence, the spectral gap between the largest and second-largest eigenvalues, the observation probability and the noise variance. We validate our theoretical results with incomplete synthetic data, and show encouraging and meaningful results on a gene expression dataset.

Related papers

Err

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.