Matrix Completion by the Principle of Parsimony
Augusto Ferrante, Michele Pavon

TL;DR
This paper extends Dempster's covariance selection method to more general matrices and demonstrates that these generalized completions solve an entropy-like variational problem, broadening the theoretical understanding of matrix completion.
Contribution
The paper generalizes Dempster's covariance selection to nonsingular and full rank rectangular matrices, linking matrix completion to entropy-like variational problems.
Findings
Generalized matrix completion methods to broader classes of matrices.
Established a connection between matrix completion and entropy-like variational problems.
Extended the theoretical framework of covariance selection.
Abstract
Dempster's covariance selection method is extended first to general nonsingular matrices and then to full rank rectangular matrices. Dempster observed that his completion solved a maximum entropy problem. We show that our generalized completions are also solutions of a suitable entropy-like variational problem.
Peer Reviews
No public reviews on file for this paper yet. If you reviewed it on a platform where reviews are public (OpenReview, ICLR, NeurIPS, ICML), you can paste yours below so the community can read it here.
Videos
No videos yet. Explain this paper in a talk, walkthrough, or lecture? Add one.
Taxonomy
TopicsStatistical Mechanics and Entropy · Topological and Geometric Data Analysis · Complex Systems and Time Series Analysis
