Subsampling for Large Graphs Using Ricci Curvature.

In the real world, there are increasingly large networks, including disease networks, contact tracing networks, brain networks, etc. Most networks have thousands or, millions of nodes or edges. However, there are some challenges in large networks: hard to visualize; hard to analyze; and hard for downstream computation; Thus, we propose a manifold-based network sampling method that does not need access to the whole graph and utilizes local features with global information.

Anomaly Detection, Diagnosis, and Localization in Smart Grid

In Smart grids, power electronics are the fundamental building blocks. A catastrophic failure of power electronic networks due to a malicious cyber-attack, or an accidental physical fault would cause degradation of equipment and substantial economic loss. This project presents a novel framework for solving anomaly detection, diagnosis, and localization problems in power electronic networks. Compared to other deep learning algorithms that need labels of the normal data or labels of both the normal and anomalous data, the proposed algorithm is unsupervised and does not need labels to train.