Wanning Pan
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Wanning Pan has a total of 11 co-authors. Here are the top co-authors by number of shared publications:
- Dezhong Yao
School of Computer Science and Technology, Huazhong University of Science and Technology, Wuhan, China
2 shared publications - Hai Jin
Huazhong University of Science and Technology
2 shared publications - Lichao Sun
Lehigh University, Bethlehem, USA
2 shared publications - Yao Wan
Huazhong University of Science and Technology, Wuhan, China
2 shared publications - Yutong Dai
Lehigh University, Bethlehem, PA, USA
2 shared publications - Chen Yu
School of Computer Science and Technology, Huazhong University of Science and Technology, Wuhan, China
1 shared publication
Latest Publications
- Open Access
- 25 April 2025
- Published by Elsevier in Artificial Intelligence
- Vol. 344
FedHM: Efficient federated learning for heterogeneous models via low-rank factorization
- 25 April 2025
- Published by Elsevier in Artificial Intelligence
- Vol. 344
Open AccessFedHM: Efficient federated learning for heterogeneous models via low-rank factorization
- 15 September 2023
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Computers
- Vol. 73 (1) , 3-17
FedGKD: Toward Heterogeneous Federated Learning via Global Knowledge Distillation
Federated learning, as one enabling technology of edge intelligence, has gained substantial attention due to its efficacy in training deep learning models without data privacy and network bandwidth concerns. However, due to the heterogeneity of the edge computing system and data, many methods suffer from the “ client-drift ” issue that could considerably impede the convergence of global model training: local models on clients can drift apart, and the aggregated model can be different from the global optimum. To tackle this issue, one intuitive idea is to guide the local model training by global teachers, i.e. , past global models, where each client learns the global knowledge from past global models via adaptive knowledge distillation techniques. Inspired by these insights, we propose a novel approach for heterogeneous federated learning, F ed GKD, which fuses the knowledge from historical global models and guides local training to alleviate the “ client-drift ” issue. In this paper, we evaluate F ed GKD through extensive experiments across various CV and NLP datasets (i.e., CIFAR-10/100, Tiny-ImageNet, AG News, SST5) under different heterogeneous settings. The proposed method is guaranteed to converge under common assumptions and outperforms the state-of-the-art baselines in the non-IID federated setting.
- 15 September 2023
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Computers
- Vol. 73 (1) , 3-17
FedGKD: Toward Heterogeneous Federated Learning via Global Knowledge Distillation
Federated learning, as one enabling technology of edge intelligence, has gained substantial attention due to its efficacy in training deep learning models without data privacy and network bandwidth concerns. However, due to the heterogeneity of the edge computing system and data, many methods suffer from the “ client-drift ” issue that could considerably impede the convergence of global model training: local models on clients can drift apart, and the aggregated model can be different from the global optimum. To tackle this issue, one intuitive idea is to guide the local model training by global teachers, i.e. , past global models, where each client learns the global knowledge from past global models via adaptive knowledge distillation techniques. Inspired by these insights, we propose a novel approach for heterogeneous federated learning, F ed GKD, which fuses the knowledge from historical global models and guides local training to alleviate the “ client-drift ” issue. In this paper, we evaluate F ed GKD through extensive experiments across various CV and NLP datasets (i.e., CIFAR-10/100, Tiny-ImageNet, AG News, SST5) under different heterogeneous settings. The proposed method is guaranteed to converge under common assumptions and outperforms the state-of-the-art baselines in the non-IID federated setting.