Dr Hongyuan ZHANG
Post-doctoral Fellow
HKU Musketeers Foundation Institute of Data Science
HKU Musketeers Foundation Institute of Data Science
Key Expertise
Representation Learning
About Me
Dr. Hongyuan Zhang is currently a Postdoctoral Fellow of The HKU Musketeers Foundation Institute of Data Science, The University of Hong Kong, working with Prof. Ping Luo. He received his Ph.D. degree from Northwestern Polytechnical University in 2024 under the supervision of Prof. Xuelong Li. He is the Associate Editors / Guest Editors of several top-tier journals including Pattern Recognition and Remote Sensing. He has published over 20 cutting-edge papers on representation learning and machine learning (e.g., T-PAMI, T-IP, CIKM, etc.). He serves as PC members/reviewers of top-tier conferences and journals, e.g., ICML, NeurIPS, ICLR, CVPR, ICCV, ECCV, AISTATS, TPAMI, etc.
Selected Publications
- Decouple Graph Neural Networks: Train Multiple Simple GNNs Simultaneously Instead of One, TPAMI
- Non-Graph Data Clustering via O(N) Bipartite Graph Convolution, TPAMI
- Embedding graph auto-encoder for graph clustering, TNNLS
- Adaptive graph auto-encoder for general data clustering, TPAMI
- Robust multi-task learning with flexible manifold constraint, TPAMI
- Matrix completion via non-convex relaxation and adaptive correlation learning, TPAMI
- Towards projected clustering with aggregated mapping, TIP
- Data Augmentation of Contrastive Learning is Estimating Positive-incentive Noise, arXiv preprint arXiv:2408.09929
- Variational Positive-incentive Noise: How Noise Benefits Models, arXiv preprint arXiv:2306.07651
Research Interests
Representation Learning, Embodied AI, Multimodal Learning