This MATLAB package includes implementations of the graph learning algorithms presented in [1]. It requires the GLL package, which implements the algorithms from [2]. The wgplot toolbox is included to generate figures. Some of the rotated Brodatz textures are also included (http://sipi.usc.edu/database/database.php?volume=rotate).
[1] Pavez, Eduardo, Hilmi E. Egilmez, and Antonio Ortega. "Learning graphs with monotone topology properties and multiple connected components." IEEE Transactions on Signal Processing 66.9 (2018): 2399-2413.
Arxiv version: https://arxiv.org/abs/1705.10934
[2] H. E. Egilmez, E. Pavez, and A. Ortega, "Graph learning from data under Laplacian and structural constraints," IEEE Journal of Selected Topics in Signal Processing, 2017.
Arxiv version: https://arxiv.org/abs/1611.05181
To install the package: (1) Download the source files. (2) Download GLL package https://github.com/STAC-USC/Graph_Learning (3) Download/install CVX (requried for learning bipartite graphs)
The demo script 'experiment_graph_learning_topology.m' shows the usage of functions used to estimate generalized graph Laplacian matrices with topology properties (sparse connected, tree, bipartite).