This work aims at recovering signals that are sparse on graphs. Compressed sensing offers techniques for signal recovery from a few linear measurements and graph Fourier analysis provides a signal representation on graph. In this paper, we leverage these two frameworks to introduce a new Lasso recovery algorithm on graphs. More precisely, we present a non-convex, non-smooth algorithm that outperforms the standard convex Lasso technique. We carry out numerical experiments on three benchmark graph datasets.
Citation / Publisher Attribution
This article is provided for educational and research purposes. More info can be found about the Signal Processing Conference (EUSIPCO), 2015 23rd European at http://www.eusipco2015.org/content/welcome-eusipco2014-nice.
X. Bresson, T. Laurent and J. von Brecht. "Enhanced Lasso Recovery." Graph European Signal Processing Conference (EUSIPCO), 2015.
First published in the Proceedings of the 23rd European Signal Processing Conference (EUSIPCO-2015) in 2015, published by EURASIP.