A Experiment details

A.1 Dataset statistics In Table 5, we provided the statistics of all datasets used in our experiments. We run all experiments on a machine with 80 Intel(R) Xeon(R) E5-2698 v4 @ 2.20GHz CPUs, and Our implementations are based on Python 3.8.10, Curves in all three plots are the higher the better. All three metrics are the higher the better. GStarX outperforms the other methods A.5 Detailed entropy-based sparsity evaluation In Section 5.2 we study whether the obtained scores by GStarX are sparse and follow [ We now provide a more detailed discussion of this study.