Cheung Yiu-ming: Solving Real-world Problems

Prof Cheung Yiu Ming — Professor Cheung Yiu-ming, Department of Computer Science, was named as a Chair Professor of the Changjiang Scholars by the Ministry of Education.

Professor Cheung Yiu-ming, Department of Computer Science, HKBU, is an outstanding scholar who has a spirit of exploration and curiosity. From his impactful patented technology “Infrared-Spectrum Based Recognition System for Identifying Chinese Herbal Species” to the novel award-winning inventions “fatigue driving detection and alarm system” and “lip-password authentication security system”, Professor Cheung has always had a relentless passion for transferring his scientific knowledge and research findings to real-life applications.

Reputation for innovation

Through his research efforts in an array of fields such as machine learning, data science, computer vision, pattern recognition, and optimisation, Professor Cheung has garnered a reputation for his many inspiring and empowering studies. In addition to being an IEEE Fellow, IET Fellow, and British Computer Society (BCS) Fellow, he was ranked in the World’s Top 1% Most-cited Scientists in the field of AI and Image Processing by Stanford University in 2020. He was also named as a Chair Professor of the Changjiang Scholars by the Ministry of Education, affiliated with State Key Laboratory of Synthetical Automation for Process Industries (SKL-SAPI) in Northeastern University (NEU) for his outstanding contributions to artificial intelligence and computer vision in 2019.

Partnership for world-class research

Fig 1: Illustration of different shifts among classes. Ct 1 and Ct 2 are the two historical concepts, and CurCt is a concept at the current time step. The drift between Ct 1 and CurCt exists in Class 1 and Class 2. The drift across the Ct 2 and CurCt only causes the labels of Class 3 and Class 4 swapping.

Professor Cheung views his Chair Professorship not only as an honour for himself, but also a benefit to HKBU. “This not only promotes research collaboration,” stated Professor Cheung, “but be also helpful for HKBU’s development as a hub for world-class research.” Recently, Professor Cheung’s team at SKL-SAPI published a paper in IEEE Transactions on Neural Networks and Learning Systems entitled “Concept Drift-tolerant Transfer Learning (CDTL) in Dynamic Environments” (DOI: 10.1109/TNNLS.2021.3054665).

In this paper, the research team proposes a hybrid ensemble approach to dealing with the CDTL problem, aiming to adapting the target model and knowledge of source domains to the dynamic environments. The proposed algorithm allows positive knowledge of source domains to be potentially promoted while negative knowledge to be reduced, thus making the transfer learning more effective.

In the long-run, the mutual collaboration track record with the SKL-SAPI at NEU could make it possible for HKBU to become a partner with SKL-SAPI and to establish a new AI-based State Key Laboratory at our campus.

Professor Cheung Yiu-ming

Department of Computer Science

Vision for the future

Despite the fact that Professor Cheung already has over 250 articles published in the high-quality conferences and journals while serving in a number of prestigious journals and esteemed bodies like the IEEE Computer Society, he is still highly devoted in supporting the University’s ambition and vision to drive research forward to the next level. “In the long-run, the mutual collaboration track record with the SKL-SAPI at NEU could make it possible for HKBU to become a partner with SKL-SAPI and to establish a new AI-based State Key Laboratory at our campus,” Professor Cheung said, “This would be a most exciting accomplishment!”

VIDEO

World’s first “lip password”: A patented double security system for identity authentication