Graph Machine Learning: Foundations and Perspectives
Abstract:
In recent years, research on analyzing and mining graph data using machine learning methods has received increasing attention due to the significant expressive power of graph-structured data. Graph Neural Networks (GNNs), a family of deep learning-based methods for processing graph data, have shown excellent performance in many fields and have become widely used for graph analysis. In this talk, we will explore key tasks and cutting-edge applications of GNNs. We will then present three of our recent research works on the scalability of spectral GNNs, temporal graph learning, and graph unlearning. Finally, we will explore two promising future directions for graph machine learning in the era of Large Language Models (LLMs): graph for science and graph for social science.
Bio:
Zhewei Wei is a Professor at the Gaoling School of Artificial Intelligence, Renmin University of China. He received his BSc from Peking University in 2008 and his Ph.D. from Hong Kong University of Science and Technology in 2012. Afterward, he worked as a Postdoc at Aarhus University, from 2012 to 2014, and then joined Renmin University of China in 2014. He has published over 80 papers in top conferences and journals (e.g., STOC, SIGMOD, ICML, KDD) in the fields of theoretical computer science, databases, machine learning, and data mining. He received the Alberto Mendelzon PODS 2022 Test of Time Award, the 2023 World Artificial Intelligence Conference Youth Outstanding Paper Nomination Award, and the VLDB 2023 Best Paper Nomination Award. He has served as an Associate Editor for IEEE TPAMI, the Proceedings Chair for PODS and ICDT, and the Area Chair for ICML, NeurIPS, and ICLR. His Ph.D. students have been awarded the Baidu Scholarship 2021 (10 worldwide), the Microsoft Scholar 2022 (12 in the Asia-Pacific region), and the CCF Outstanding Doctorial Dissertation Award (10 nationwide).