This tutorial covers several fontiers, including graph generation, self-supervised learning of GNNs, explainability of GNNs and 3D GNNs. We firstly introduce graph neural networks briefly. Then, for each of the mentioned directions, we provide a unified and comprehensive review of the key approaches. In addition, we have hands-on code demonstrations based on our library DIG to help the audience to implement and benchmark methods in these frontiers without extra effort.
Shuiwang Ji is currently a Professor and Presidential Impact Fellow in the Department of Computer Science and Engineering, Texas A&M University. His research interests include artificial intelligence, machine learning, and graph analysis.
Meng Liu is a third-year Ph.D. student in the Department of Computer Science and Engineering, Texas A&M University. Currently, he is working on graph deep learning, AI for science, and generative modeling.
Yi Liu is a Ph.D. Candidate in the Department of Computer Science and Engineering, Texas A&M University. His research interests include artificial intelligence, machine learning, and quantum information science.
Youzhi Luo is a third-year Ph.D. student in the Department of Computer Science and Engineering, Texas A&M University. His current research interests are graph neural networks and deep generative models, and their applications to molecular property prediction and molecule generation.
Limei Wang is a third-year Ph.D. student in the Department of Computer Science and Engineering, Texas A&M University. Currently, she is working on graph representation models and their applications to molecular property prediction and drug discovery.
Yaochen Xie is a fourth-year Ph.D. student in the Department of Computer Science and Engineering, Texas A&M University. His research interests include self-supervised learning, graph data mining, and explainability of graph neural networks.
Zhao Xu is a third-year Ph.D. student in the Department of Computer Science and Engineering, Texas A&M University. Her research interests include molecular geometry and property prediction, and self-supervised learning.
Haiyang Yu is a second-year Ph.D. student in the Department of Computer Science and Engineering, Texas A&M University. His research interests includes graph deep learning, explainability, and AI for molecular property prediction.