Imbalanced Data Learning: from Class Imbalance to Long-tailed Data Classification for Visual Recognition

Imbalanced Data Learning: from Class Imbalance to Long-tailed Data Classification for Visual Recognition

2025 UTS FEIT Research Excellence Guest Lecture Series - Prof. Yiu-ming Cheung

Abstract
Imbalanced data refers to the number of samples among classes is extremely imbalanced, which is common in our daily life, e.g. medical diagnosis, and autonomous driving. In general, the problem of learning from imbalanced data is nontrivial and challenging in the field of data engineering and machine learning, which has attracted growing attention in recent years. In this talk, the imbalance data learning problem is introduced from class imbalance to long-tailed data learning, including their potential applications, and the impacts from a model learning perspective. Then, the latest research progress on imbalanced data learning will be reviewed, including some representative methods in the literature. Lastly, the potential research directions in this field will be discussed.

Speaker Biography
Prof. Yiu-ming Cheung is Chair Professor in the Department of Computer Science at Hong Kong Baptist University, where he also serves as Dean of the Institute for Research and Continuing Education and Associate Director of the Institute of Computational and Theoretical Studies. He is a Fellow of IEEE, AAAS, IAPR, IET, BCS, and a member of the European Academy of Sciences and Arts.

Recognised among the world’s top 1% most-cited scientists in AI and image processing by Stanford University since 2019, Prof. Cheung’s research spans machine learning, visual computing, data science, and information security. His innovations have led to multiple patents and award-winning technologies, including a lip-password for identity verification that earned top honours at international invention exhibitions in Geneva and the Middle East. He is currently Editor-in-Chief of IEEE Transactions on Emerging Topics in Computational Intelligence and serves on editorial boards of several leading journals. Prof. Cheung also plays a key role in the global AI community as Founding Chair of IEEE (HK) Computational Intelligence Chapter and Chair of IEEE Technical Community on Intelligent Informatics.