Title :
Matrix Completion à la Dempster by the Principle of Parsimony
Author :
Ferrante, Augusto ; Pavon, Michele
Author_Institution :
Dipt. di Ing. dell´´Inf., Univ. di Padova, Padova, Italy
fDate :
6/1/2011 12:00:00 AM
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.
Keywords :
covariance matrices; maximum entropy methods; Dempster covariance selection method; full rank rectangular matrices; matrix completion; maximum entropy problem; nonsingular matrices; principle of parsimony; Covariance matrix; Entropy; Equations; Indexes; Linear systems; Stacking; Symmetric matrices; Covariance selection; matrix completion; maximum entropy problem; parsimony principle; variational problem;
Journal_Title :
Information Theory, IEEE Transactions on
DOI :
10.1109/TIT.2011.2143970
