Real-time critical transition discoveries with large language models
Guijun Ma, Zidong Wang, Yuzhe Wang, Yong Zhang, Haitao Song, Han Ding, Ye Yuan
National Science Open, 5(2): 20250060, 2026
Guijun Ma (Marek)
Physical AI Lab, School of Artificial Intelligence and Automation
Huazhong University of Science and Technology (HUST)
I develop learning methods for intelligent systems that interact with the physical world. My research connects Physical AI and transfer learning, with applications in generalized robot learning and intelligent manufacturing.
My goal is to enable robots and industrial systems to recognize hidden physical states, learn transferable dynamics, and make reliable decisions under limited data and changing environments.
Email: mgj@hust.edu.cn
Office: School of Artificial Intelligence and Automation, HUST, Wuhan, China
Research Interests
Physical AI & Safe Generalization
Physics-informed learning, latent-state modeling, uncertainty, and reliable adaptation under distribution shift.
Robot Learning & Manipulation
Contact-rich manipulation, visual–tactile learning, mobile manipulation, and cross-task skill transfer.
Robotized Smart Manufacturing
Robotic machining, process prediction, calibration, and closed-loop optimization.
Battery Intelligence
State-of-health estimation, remaining-useful-life prediction, and personalized battery prognostics.
News
- Our work “Real-time critical transition discoveries with large language models” was selected as an NSO cover article.
- Our work “Learning linear state-space models with sparse system matrices” was accepted by ICLR 2026.
- Our work “Closed-Loop Parameter Optimization for Robotic Machining Using Physics-Informed Machine Learning and Multiobjective Optimization” appeared in IEEE T-ASE.
- Our work “A Novel Pairwise Domain-Adaptation-Assisted Dual-Task Learning Approach to Coprediction of Robotic Machining Efficiency and Quality in New Parameter Spaces” appeared in IEEE TII.
- Our work “A Transfer Learning-Based Method for Personalized State of Health Estimation of Lithium-Ion Batteries” appeared in IEEE TNNLS.
- Our work “Estimating the State of Health for Lithium-ion Batteries: A Particle Swarm Optimization-Assisted Deep Domain Adaptation Approach” appeared in IEEE-CAA JAS.
- Our work “Real-time personalized health status prediction of lithium-ion batteries using deep transfer learning” was selected as an EES cover article.
Selected Publications
Full list on Google Scholar →
Closed-Loop Parameter Optimization for Robotic Machining Using Physics-Informed Machine Learning and Multiobjective Optimization
Guijun Ma, Zidong Wang, Weibo Liu, Zeyuan Yang, Desheng Huang, Han Ding
IEEE Transactions on Automation Science and Engineering, 22: 22410–22422, 2025
Real-Time Personalized Health Status Prediction of Lithium-Ion Batteries Using Deep Transfer Learning
Guijun Ma, Songpei Xu, Benben Jiang, Cheng Cheng, Xin Yang, Yue Shen, Tao Yang, Yunhui Huang, Han Ding, Ye Yuan
Energy & Environmental Science, 15: 4083–4094, 2022
Featured Covers
Selected journal features
National Science Open · Cover Article
Real-Time Critical Transition Discoveries with Large Language Models
2026
Energy & Environmental Science · Cover Article
Real-Time Personalized Health Status Prediction of Lithium-Ion Batteries Using Deep Transfer Learning
2022
More to come
More research highlights coming soon.
Future cover features, best-paper recognition, and selected research stories will appear here.
Biography
Guijun Ma has been a Lecturer at the School of Artificial Intelligence and Automation, Huazhong University of Science and Technology (HUST), since Apr. 2026.
Prior to his current appointment, he worked as a Postdoctoral Researcher at HUST from 2023 to 2025. He received his B.Eng. degree and Ph.D. degree from HUST in 2017 and 2023, respectively. His doctoral research was jointly supervised by Prof. Han Ding and Prof. Ye Yuan. From 2021 to 2022, he was a Visiting Scholar in the Department of Computer Science at Brunel University London, supervised by Prof. Zidong Wang.
He currently serves as an Associate Editor of Electronics Letters. He has received the Best Paper Award from National Science Review and consecutive Best Paper Awards at the International Conference on Advanced Control and Artificial Intelligence (ICACAI) in 2025 and 2026.
He welcomes international collaboration on physical intelligence, transferable robot skills, visual–tactile manipulation, mobile manipulation in confined environments, and trustworthy industrial AI.