Ongoing Research Projects and Activities:

Project 1: Explainability in Graph Neural Networks

The pervasiveness of machine learning solutions in critical domains such as autonomous vehicles, medical diagnostics, financial systems, and even sensitive security and military applications has created an urgent need for trustworthy and explainable AI (XAI). The black-box nature of many state-of-the-art machine learning models, such as DNNs and GNNs, has become a significant concern for both practitioners and regulators. In this project, we develop an information theoretic framework for GNN explainability along with explanation methods and implementable algorithms... Read More

Project 2: An Information Theoretic Framework for Network Privacy

Internet users reasonably expect their online identities and web browsing activities to remain private. Unfortunately, this is far from the case in practice;in reality, users are constantly tracked on the internet.  As web tracking... Read More