Home Page

Papers

Submissions

News

Scope

Editorial Board

Announcements

Proceedings

Open Source Software

Search

Login



RSS Feed

The Bigraphical Lasso

Alfredo Kalaitzis, John Lafferty, Neil Lawrence, Shuheng Zhou
;
JMLR W&CP 28 (3) : 1229–1237, 2013

Abstract

The i.i.d. assumption in machine learning is endemic, but often flawed. Complex data sets exhibit partial correlations between both instances and features. A model specifying both types of correlation can have a number of parameters that scales quadratically with the number of features and data points. We introduce the bigraphical lasso, an estimator for precision matrices of matrix-normals based on the Cartesian product of graphs. A prominent product in spectral graph theory, this structure has appealing properties for regression, enhanced sparsity and interpretability. To deal with the parameter explosion we introduce L1 penalties and fit the model through a flip-flop algorithm that results in a linear number of lasso regressions.

Related Material