| Speaker | Yuchen Lin, University of Southern California |
| Host | Zhongyu Wei, Fudan University |
| Time | 2020-6-27 11:00-12:30 |
| Zoom ID | 639 9966 4124 |
| 4124Password | 123456 |
| Abstract | Can you answer this multi-choice question: What do you need to fill with ink for writing in a notebook? A: a pencil, B: a printer, C: a pen. This is a trivial question for humans, even for kids. Can we make machines also capable of doing such inferences using basic knowledge about the world? This research question is called commonsense reasoning, which has plagued the field of artificial intelligence for over 60 years. There have been two main types of approaches to empower machines with commonsense reasoning ability: symbolic systems and neural networks. |
| Bio | Yuchen Lin is currently a Ph.D. student at the University of Southern California, supervised by Prof. Xiang Ren. Previously, he got his bachelor's degree at the IEEE Honor Class of Shanghai Jiao Tong University. He is now doing a summer intern at Google AI, in Dr. William W. Cohen’s team. He is generally interested in building neural-symbolic systems that demonstrate a deep understanding of the world, integrating technologies for information extraction, knowledge graphs, machine reasoning, graph neural networks, and model robustness. |