- Jan 2026: Our paper on Intra-Cluster Mixup: An Effective Data Augmentation Technique for Complementary-Label Learning has been accepted to Transaction of Machine Learning Research (TMLR).
- May 2025: Our paper on CLImage: Human-Annotated Datasets for Complementary-Label Learning has been accepted to Transaction of Machine Learning Research (TMLR).
- Feb 2025: Our paper on The Unexplored Potential of Vision-Language Models for Generating Large-Scale Complementary-Label Learning Data has been accepted to PAKDD 2025.
Mai Tan Ha (Ph.D. Candidate) [ GitHub | Google Scholar ]
Researcher (a.k.a. PhD Candidate, Research Assistant at CLLab)Computer Science and Information Engineering, National Taiwan University
Advisor: Prof. Hsuan-Tien Lin
Visiting Scholar
RIKEN Center for Advanced Intelligence Project (RIKEN AIP), Imperfect Information Learning Team
RIKEN Advisors: Prof. Masashi Sugiyama and Dr. Gang Niu
Email: maitanhaksdtvt6@gmail.com or d10922024@ntu.edu.tw
Research interests: Machine Learning, Artificial Intelligence, Data Mining.
I'm a machine-learning researcher specializing in weakly supervised and imbalanced learning. My current work focuses on making complementary-label learning both realistic and robust. My ultimate goal is to become a lecturer who empowers and inspires the next generation of innovators. Outside the lab, you'll find me on the football pitch or hiking through the forest on a weekend camping trip.