Graph neural networks (GNNs) have emerged as a powerful framework for learning from graph-structured data, yet their theoretical understanding—particularly regarding the behavior of different architectural choices across various graph-based tasks—remains limited. In parallel, random geometric graphs (RGGs) provide a well-defined probabilistic model that captures the interplay between geometry and connectivity in complex networks.
In this talk, I will discuss several efforts I have undertaken to bridge these two perspectives by studying GNNs through the lens of RGGs. In the first part, I will focus on the classic graph matching problem and show that, by leveraging a specific GNN, perfect recovery can be achieved even in high-noise regimes. In the second part, I will briefly highlight recent work demonstrating the provable benefits of graph attention networks (GATs) for a node regression task. This talk is based on joint work with Morgane Austern, Kenny Gu, and Somak Laha.
The talk will be given by Prof. Suqi Liu, Department of Statistics, UCR
Suqi Liu is an Assistant Professor in the Department of Statistics at the University of California, Riverside. He received his Ph.D. in Operations Research and Financial Engineering from Princeton University. Prior to joining UCR, he was a Postdoctoral Research Fellow in Biomedical Informatics at Harvard University. His research interests lie at the intersection of probability, statistics, and artificial intelligence, with applications in biomedical sciences.