Here are some of the recent invited talks that I have given:
Explainability in Graph Machine Learning
- Invited tutorial at XAISS summer school 2022
- Date: 08/30/2022
- Description: The tutorial covers the recent advances in the area of post-hoc explanations for graph neural networks. In addition we discuss the challenges of evaluating post-hoc eplanations and the recent benchmarks. A hands-on session was also organized.
Privacy in GNNs
- Invited talk at NISER, India
- Date: 04/21/2022
- Description: Graph Neural Networks (GNNs), which generalize traditional deep neural networks on graph data, have achieved state-of-the-art performance on several graph analytical tasks. With their growing popularity in various applications including sensitive applications like health it is imperative to study the privacy aspects of these models. In this talk I discussed the vulnerability of GNNs to privacy leakage and also present our framework on releasing GNNs under differential privacy guarantees.
Opportunities and challenges in Graph Machine Learning for Biomedicine
- Invited talk at LeibnizAILab, Germany
- Date: 10/26/2022
- Description: Graph Machine Learning (GraphML), whereby classical machine learning is generalized to irregular graph domains, has enjoyed a recent renaissance, leading to a dizzying array of models and their applications in several domains. In this talk, I provide an overview of our recent works in which we exploited GraphML to achieve joint learning on multiple biological information sources to build predictive and generalizable models. Besides, I discuss briefly the challenges of privacy and transparency as well as their interplay in GraphML, which could hinder the deployment of such models in actual clinical settings.