About 中文

I’m now an Assistant Professor and PhD supervisor in Department of Computer Science and Technology, School of Informatics, Xiamen University. I received my PhD degree on computer science from Hong Kong Baptist University in 2019 supervised by Prof. Yiu-ming Cheung and received my bachelor and master degree on software engineering from University of Macau in 2012 and 2014, supervised by Prof. Yuan Yan Tang. I was a research intern at Tencent in 2018. I have published over 40 top-tier papers, many of which have been featured in top-tier machine learning journals (JCR Q1) such as IEEE TNNLS and IEEE TCYB, as well as premier AI and computer vision conferences (CCF-A) like CVPR, ICCV, AAAI, and IJCAI. I have led several projects, including the General Project and Youth Project of the National Natural Science Foundation of China, the Major Project of the Fujian New Think Tank, the General Project of the Fujian Natural Science Foundation, and open projects at the Zhejiang Lab. I serve as a member of the Machine Vision Committee and the Secretary-General of the Xiamen Member Activity Center of the China Society of Image and Graphics. I also serve as the Associate Editor of the international journal IEEE Transactions on Emerging Topics in Computational Intelligence. I was honored as the “Xiaomi Young Scholar” by Xiamen University in 2023. My research interest includes artificial intelligence, machine learning. My research direction is trustworthy deep learning, including long-tail learning, federated learning, label-noise learning and continual learning.

For undergraduate students, please check 本科招生简章. For postgraduate students, please check 硕士招生简章.

There is still a few postgraduate quota in Fall 2025. Please contact me ASAP.

Research Interests

Deep Learning with Incomplete Labels

Machine learning with incomplete labels addresses effective model training in the presence of significant label deficiencies, such as imbalanced label distribution or noisy labels. The goal is to achieve accurate predictions despite low data quality. We enhance model performance on incomplete labeled data by introducing more effective label processing and integration strategies.

Long-tailed Visual Recognition (CVPR’23, CVPR’22, IJCAI’22, AAAI’24, AAAI’24)
Imbalance Learning (IJCAI’24, IJCAI’17, TNNLS’20, TNNLS’20, TCYB’21)
Label-Noise Learning (ICCV’23, AAAI’24, TCYB’23, ECML-PKDD’21)

Trustworthy Federated Learning

Trustworthy federated learning aims to perform model training in a distributed data environment, ensuring data privacy and security while maintaining high model performance. It also explores personalization and fairness within the federated learning framework. Our work focuses on data heterogeneity, long-tailed distribution, missing labels, and noise issues, improving the robustness and performance of models in heterogeneous data and noisy environments.

Long-tailed Federated Learning (IJCAI’22, AAAI’24, ICME’22, ICASSP’23)
Extreme Label-Noise in Federated Learning (AAAI’24)
Semi-supervised Federated Learning (Arxiv’23)

Continual Learning and Prompt Learning

In real-world applications, the data environment is dynamically changing, and models need to continuously learn new knowledge while retaining existing knowledge without forgetting. Our research involves task-imbalanced continual learning, concept drift handling in incremental learning, and domain generalization, aiming to enhance model stability and adaptability in dynamic and imbalanced data environments.

Task-Imbalanced Continual Learning (IJCAI’24)
Incremental Learning with Concept Drift (IJCAI’17, TNNLS’20)
Vision-Language Prompt Learning (Arxiv’24)

Supervised Students

Pinxin Qian (2020-2023, Employed in Alibaba): Federated learning
Xinyi Shang (2020-2023, Pursuing PhD at UCL): Federated learning
Yiliang Zhang (2021-2024, Employed in Alibaba): Label-noise learning
Yan Jin (2021-2024, Pursuing PhD at BUAA): Long-tail learning
Xiaolin Huang (2021-2024, Employed in JD): Semi-supervised learning
Fengling Lv (2021-2024, Employed in Xiaomi): Federated learning
Chengbin Zheng (2021-2024, Employed in CATL): Graph neural networks
Chenxing Hong (2022-): Continual learning
Lin Chen (2022-): Federated learning
Liyuan Wang (2022-): Long-tail learning
Yijie Liu (2023-): Federated learning
Shanshan Yan (2023-): Federated learning
Yizhou Chen (2023-): Semi-supervised learning
Hezhao Liu (2023-): Continual learning
Shu Chen (2023-): Label-noise learning
Zenghui Huang (2024-): Long-tail learning
Ruichi Zhang (2024-): Long-tail learning
Junpeng Chen (2024-): Long-tail learning
Shihao Hou (2024-): Federated learning

Yang Lu