Title of the Report:Intelligent Industry and Scientific Machine Learning

Speaker:Yuntian Chen

Affiliation:East China University of Science and Technology, Ningbo

Date and Time:April 11, 2024 (Thursday), 9:30-10:30

Location:Room 5, 1st Floor, Block A, Jade Science and Education Building

Abstract:

Machine learning has been widely applied in the industrial domain, but it still faces challenges such as insufficient accuracy and robustness of models due to data scarcity and complex scenarios. This report explores the integration of industrial knowledge and machine learning models, aiming to enhance the accuracy and robustness of models by embedding knowledge and discovering knowledge in scientific machine learning, thereby achieving a closed loop between knowledge and data in the intelligent industrial domain. Knowledge embedding can break down barriers between knowledge and data, thus establishing machine learning models with physical common sense. Human understanding of the world is always limited, and knowledge discovery can utilize machine learning to extract new knowledge from observations. Knowledge discovery not only helps researchers better understand the essence of physics but also provides favorable support for the research of knowledge embedding. By combining knowledge embedding and knowledge discovery, a closed-loop of knowledge generation and utilization in the industrial domain can be formed, thereby constructing intelligent industrial models that are physically reasonable, mathematically accurate, and computationally stable and efficient.

Speaker's Bio:

Yuntian Chen is an Associate Professor (Associate Professor, Ph.D. Supervisor) at East China University of Science and Technology, Ningbo, and also serves as a Ph.D. Supervisor in the Department of Computer Science at Shanghai Jiao Tong University. His research focuses on scientific machine learning, primarily in the areas of 1) Fusion of physics-driven and data-driven approaches and 2) AI-driven discovery of physical laws. He graduated from the Department of Energy and Power Engineering at Tsinghua University with a dual degree in Economics from Peking University. He completed his Ph.D. studies at the School of Engineering, Peking University, and was honored as an outstanding graduate. He was a postdoctoral fellow at Peng Cheng Laboratory. He has published over 30 papers, been granted 19 invention patents, and led 10 projects funded by the National Natural Science Foundation of China, with a cumulative funding of over 15 million RMB. He has received honors such as the Top Paper Award from a prestigious Chinese journal and the Gold Award in the China "Internet Plus" Innovation and Entrepreneurship Competition. He has been selected for the Young Talent Reservoir Plan of the Chinese Society of Mechanics and the Yongjiang Talent Project.