KEYNOTE SPEAKERS  

Prof. Qingzu Zhu
University of Science and Technology of China, China

Speech Title: TBA
Abstract: TBA

Biography: Zuqing Zhu received his Ph.D. degree from the Department of Electrical and Computer Engineering, University of California, Davis, in 2007. From 2007 to 2011, he worked in the Service Provider Technology Group of Cisco Systems, San Jose, California, as a Senior Engineer. In January 2011, he joined the University of Science and Technology of China, where he currently is a Full Professor in the School of Information Science and Technology. He has published 360+ papers in peer-reviewed journals and conferences. He is the Steering Committee Chair of the IEEE International Conference on High Performance Switching and Routing (HPSR), and was the Chair of the Technical Committee on Optical Networking (ONTC) in IEEE Communications Society. He has received the Best Paper Awards from ICC 2013, GLOBECOM 2013, ICNC 2014, ICC 2015, ONDM 2018, and GLOBECOM 2023. He is a Fellow of IEEE.

 

Prof. Qingzu Zhu
University of Science and Technology of China, China

Speech Title: Challenges in Building Models with Less Data or Without Labels: Domain Generalization and Anomaly Detection
Abstract: Developing deep learning models with limited, unlabeled, or no data presents substantial challenges, especially in domains such as medicine, mechanical systems, and environmental monitoring. In medicine, obtaining labeled data requires expert annotation, which is costly, time-consuming, and often restricted by ethical and privacy regulations. Mechanical systems pose a different challenge, as critical failure events are rare, leading to sparse and imbalanced datasets. Environmental applications, such as UAV-based anomaly detection on water and soil conservation facilities, face logistical difficulties in label collection due to complex terrain and limited expert access. These constraints hinder effective training, increase the risk of overfitting, and weaken generalization to unseen data. This talk introduces three approaches to address these challenges: Unsupervised Domain Adaptation, Domain-Generalized Semantic Segmentation, and One-Class Anomaly Detection.

Biography: Hung-Hsu Tsai received the BS and the MS degrees in applied mathematics from National Chung Hsing University, Taichung, Taiwan, in 1986 and 1988, respectively, and the PhD degree in computer science and information engineering from National Chung Cheng University, Chiayi, Taiwan, in 1999. Currently, he is the Dean of School of Informatics and Data Science, since Feb., 2025, the Vice Dean of College of Science since Feb., 2024, and a professor at Graduate Institute of Data Science and Information Computing / Department of Applied Mathematics, National Chung Hsing University, Taichung, Taiwan. His research interests include artifical intelligence, machine learning, deep learning, multimedia watermarking, intelligent filter design, content-based multimedia retrieval.

CO-SPONSORED BY
SUPPORTED BY