Recently, Xu Zihuan, a master’s student from the Mechatronic and Vehicle Engineering School of Chongqing Jiaotong University (supervisor: Zhang Xia), published a research paper titled “A stochastic configuration network based on an improved sampling strategy and Bayesian optimization” in the Chinese Academy of Sciences (CAS) Zone 1 Top Journal Knowledge-Based Systems. The stochastic configuration network (SCN) algorithm proposed in the paper, which integrates an improved sampling strategy and Bayesian optimization, addresses the core issues of traditional SCN, such as reduced model performance and reliance on manual parameter tuning when handling complex tasks, and breaks through the performance bottleneck of traditional SCN in such tasks.

Inspired by stochastic sampling strategies and weight initialization theories, this paper is a research achievement in the field of neural networks regarding supervised stochastic learning algorithms.

The Mechatronic and Vehicle Engineering School is committed to promoting the academic innovation and scientific research capabilities of postgraduate students, regularly organizing academic activities such as the “Red May for Intelligent Innovation” Postgraduate Forum and the Postgraduate Autumn Academic Series, creating a strong research atmosphere, and continuously advancing academic programs to support more postgraduate students achieve further success in their academic careers.