On November 10, the seminar series of 100 International Scholars Seminar Report from the Faculty of Computing was held online. Invited by Professor Ma Ying of the Faculty of Computing, Researcher Yan Ming from the Advanced Artificial Intelligence Research Center of Singapore Science and Technology Development Council brought an academic report with the theme of Low resource Natural Language Understanding to teachers and students.

Yan Ming is currently a scientist at the Advanced Artificial Intelligence Research Center of Singapore Science and Technology Development Council. He has published more than 10 papers in international top conferences and journals, including ACL, IEEE TNNLS, IEEE TII, KBS, and Neural Computing. He has won 5 national patents, and has undertaken and participated in more than 10 national science and technology research and development projects in China and Singapore.

Yan Ming first gives a brief introduction to the task of natural language understanding, and also introduces the low resource problems and challenges that widely exist in the task of natural language understanding. In order to solve the performance loss and data dependence caused by low resources in natural language understanding tasks, the latest research results of his research group are introduced from the perspectives of meta learning and multi task learning. In the meta learning method, the multi domain meta learning method is introduced to improve the generalization ability of the model and the improvement method on the low resource compression model. In the aspect of multi task learning, the recent research results of his research group on distillation model and quantitative model are introduced. At the same time, the Advanced Artificial Intelligence Research Center of Singapore Science and Technology Development Agency (A * STAR) is briefly introduced.