2024 XJTLU System Identification and Machine Learning Forum Successfully Concluded

13 Nov 2024

On November 2, 2024, the Department of Foundational Mathematics at Xi'an Jiaotong-Liverpool University successfully hosted the 2024 XJTLU System Identification and Machine Learning Forum. The forum aimed to provide a platform for experts and scholars in the field of system identification to exchange the latest research findings and discuss academic issues, thereby promoting scientific research cooperation and technological innovation.

 

 

During the half-day forum, Dr Feiyan Chen invited Professor Jin Guo from the University of Science and Technology Beijing, Professor Chengpu Yu from Beijing Institute of Technology, and Professor Biqiang Mu from the Academy of Mathematics and Systems Sciences, Chinese Academy of Sciences, to deliver fascinating keynote speeches on their respective research areas.

Professor Jin Guo from the University of Science and Technology Beijing commenced with a presentation titled "System Identification for Resource Optimization and Data Security," delving into the challenges and solutions associated with system identification in network environments. Professor Guo's presentation covered the latest advancements in system identification technology and shared preliminary application research findings.

Following this, Professor Chengpu Yu from Beijing Institute of Technology presented "Multi-agent Network System Identification and Optimal Control Inversion." Professor Yu's research focuses on exploring the simultaneous inversion of topology and intention under optimal management, as well as the discovery of nonlinear network systems with known topology. His presentation detailed how to use optimal control conditions to invert control objective functions and intent states and how to implement distributed state estimation and stable agent system identification under the EM framework.

Concluding the forum, Professor Biqiang Mu from the Academy of Mathematics and Systems Sciences, Chinese Academy of Sciences, shared his research on "Efficient and High Precision Identification of Stochastic Complex Systems." Associate Researcher Mou's presentation discussed the quality of identification methods from the perspective of mean square error and proposed innovative identification methods to eliminate biases that classical identification methods could not completely eliminate.

After each presentation, the audience had the opportunity to interact with the speakers, ask questions, and engage in discussions. The forum provided a valuable opportunity for participants to exchange ideas and share insights, inspiring deeper research and exploration of practical solutions.

 

 

Dr Feiyan Chen, who organised the forum, said, "System identification, as a cornerstone in numerous fields such as network information security, multi-agent systems, and complex stochastic systems, is of great significance for driving the development of related technologies."

 

By Qinru Liu

Photos courtesy of Feiyan Chen

 

13 Nov 2024