Discriminative Graph Learning with Sparsity Regularization

The code is to implement the Discriminative Graphical Lasso (Disc-GLasso) algorithm proposed in [1]. It is written and has been tested on MATLAB R2013b.

The main function is discriGLasso.m. A demo which generates synthetic data and experimental results reported in [1] is also provided in icassp17_exp.m.

If you have any questions, please contact jiunyuka at usc dot edu.

References:

[1] J.-Y. Kao, D. Tian, H. Mansour, A. Ortega and A. Vetro, "Disc-GLasso: Discriminative Graph Learning with Sparsity Regularization", ICASSP, New Orleans, USA, 2017.