Abstract

Combining perceptual learning and logic inference/symbolic reasoning has been a long standing goal of AI. Graph neural networks are powerful representation learning tools for graph data, including but not limited to social networks, molecular graphs and knowledge graphs. A natural question is whether graph neural networks hold the potential to combine perceptual learning and logic inference, and lead to the next turning point of AI. So far, the majority of research works in graph neural networks have been focused on designing neural architectures for specific applications. Concrete arguments are still missing for why graph neural networks may be the bridge between perception learning and logic inference. In this talk, I will provide two pieces of evidence, one on inductive logic programming and the other on lifted logic inference, and show that graph neural networks lead to new, efficient and effective algorithms for addressing these challenging problems. These results also suggest that graph neural networks indeed have the potential to advance AI to the next stage.

Biography

Le Song is an Associate Professor in the College of Computing, and an Associate Director of the Center for Machine Learning, Georgia Institute of Technology. He is also a Principal Engineer in Ant Financial, Alibaba. Before he joined Georgia Institute of Technology in 2011, he was postdoc in the Department of Machine Learning, Carnegie Mellon University, and a research scientist at Google. His principal research direction is machine learning, especially kernel methods and deep learning, probabilistic graphical models, and relational data modeling. He is the recipient of the NSF CAREER Award’14, and many best paper awards, including the NIPS’17 Materials Science Workshop Best Paper Award, the Recsys’16 Deep Learning Workshop Best Paper Award, AISTATS’16 Best Student Paper Award, IPDPS’15 Best Paper Award, , NIPS’13 Outstanding Paper Award, and ICML’10 Best Paper Award. He served as the area chair or senior program committee for many leading machine learning and AI conferences such as ICML, NIPS, AISTATS, AAAI and IJCAI, and the action editor for JMLR and IEEE TPAMI.