Jiaqi Ma presentation

"Towards Trustworthy Machine Learning on Graph Data"

Abstract: Machine learning on graph data (a.k.a. graph machine learning) has attracted tremendous attention from both academia and industry, with many successful applications ranging from social recommendation to traffic forecasting, even including high-stake scenarios. However, despite the huge empircal success in common cases, popular graph machine learning models often have degraded performance in certain conditions. Given the complexity and diversity of real-world graph data, it is crucial to understand and optimize the model behaviors in specific contexts.

In this talk, I will introduce my recent work on analyzing the robustness and fairness of graph neural networks (GNNs). In the first part of the talk, I will show that existing GNNs could suffer from model misspecification, due to an implicit conditional independence assumption. This observation motivates our design of a copula-based learning framework that improves upon many existing GNNs. In the second part, I will go beyond average model performance and investigate the fairness of GNNs. Through a generalization analysis on GNNs, I will show that there is a predictable disparity in GNN performance among different subgroups of test nodes. I will also discuss potential mitigation strategies.

Bio: Jiaqi Ma is a PhD candidate in School of Information at the University of Michigan. His research interests lie in machine learning and data mining. He has worked in the areas of graph machine learning, multi-task learning, learning-to-rank, and recommender systems during his PhD study and internships with Google Brain. His work has been published in top AI journals and conferences, including JMLR, ICLR, NeurIPS, KDD, WWW, AISTATS, etc. Prior to UMich, he got his BEng degree from Tsinghua University.

Contact Christine Hopper with questions.