The Machine Learning and Computational Intelligence Research Group at the School of Mathematics and Information Science, Hebei University, has published its latest research findings in the field of machine learning

Recently, the Machine Learning and Computational Intelligence Research Group from the School of Mathematics and Information Science at our university achieved significant progress in the field of machine learning. Their latest work, titled “Robust Least Squares Twin Support Vector Machine With Adaptive Pinball Loss,” has been accepted and published online in the top artificial intelligence journal IEEE Transactions on Neural Networks and Learning Systems (2026, 5-year impact factor: 11.1). Professor Xing Hongjie is the first author and corresponding author, Dr. Wang Yidan serves as co-corresponding author, and doctoral student Wang Siwei is the second author.

This study explores robust loss functions in the field of machine learning, designs a novel adaptive pinball loss function, and incorporates it into the least squares twin support vector machine, thereby proposing the robust least squares twin support vector machine. Theoretically, the study proves its noise insensitivity and algorithmic convergence, and interprets its objective function from the perspectives of within-class scatter and commission error. Experimental results demonstrate that the model achieves efficient solving speed and excellent noise resistance.

IEEE Transactions on Neural Networks and Learning Systems was established in 1990 by the Institute of Electrical and Electronics Engineers (IEEE), with its predecessor being IEEE Transactions on Neural Networks. It is a Chinese Academy of Sciences Q1 TOP journal, with an acceptance rate of 17.2%. The journal primarily publishes the latest research findings related to neural networks and learning systems, focusing on the theory and applications of neural networks, deep learning, reinforcement learning, and learning systems, covering fields such as computer vision, natural language processing, and intelligent systems.

This research work was supported by the National Natural Science Foundation of China and the Natural Science Foundation of Hebei Province.

Paper link: https://ieeexplore.ieee.org/document/11343920