Shanfeng Wang
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Shanfeng Wang has a total of 41 co-authors. Here are the top co-authors by number of shared publications:
- Maoguo Gong
Xidian University, Xi'an, China
12 shared publications - Jianzhao Li
Guangzhou Institute of Technology, Xidian University, Guangzhou, China
6 shared publications - Yu Zhou
School of Electronic Engineering, Key Laboratory of Collaborative Intelligence Systems, Ministry of Education, Xidian University, Xian, Shaanxi, China
3 shared publications - Zhao Wang
School of Electronic Engineering, Xidian University, Xian, China
3 shared publications - Ke Pan
School of Cyber Engineering, Key Laboratory of Collaborative Intelligence Systems, Ministry of Education, Xidian University, Xian, Shaanxi Province, China
2 shared publications Lanyu Yao
School of Cyber Engineering, Key Laboratory of Collaborative Intelligence Systems, Ministry of Education, Xidian University, Xian, Shaanxi Province, China
2 shared publications
Latest Publications
- 5 June 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Circuits and Systems for Video Technology
- Vol. PP (10518215) , 1
Unifying Personalization and Generalization in Federated Few-Shot Learning via Complementary Collaboration
Personalized federated few-shot learning offers a promising approach to develop customized models for each client while respecting privacy constraints and decentralized data, addressing the challenge of limited labeled sample. However, existing methods prioritize strong personalization for seen classes under extreme data scarcity, but they struggle with weaker generalization to unseen classes under non-IID conditions, which severely restrict their practical utility. To circumvent this problem, we propose a novel personalized federated few-shot learning framework that integrates two levels of complementary learning to simultaneously improve personalization and generalization in such challenging settings. At the intra-client level, a dual-branch architecture is proposed with a global branch capturing holistic image semantics and a local branch focusing on fine-grained features from random crops. These branches interact via knowledge distillation and leverage the synergy between global context and local details to enable robust representation learning from scarce data. At the inter-client level, we propose a collaborative aggregation strategy that prioritizes partners with maximally complementary feature distributions. By integrating knowledge from both similar and diverse clients quantified by maximum mean discrepancy-based prototype discrepancy, our approach effectively leverages cross-federation information. This dual-level strategy ensures strong personalization for seen classes and robust generalization to unseen classes. Experiments on benchmark datasets confirm the effectiveness of our proposed method.
- 5 June 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Circuits and Systems for Video Technology
- Vol. PP (10518215) , 1
Unifying Personalization and Generalization in Federated Few-Shot Learning via Complementary Collaboration
Personalized federated few-shot learning offers a promising approach to develop customized models for each client while respecting privacy constraints and decentralized data, addressing the challenge of limited labeled sample. However, existing methods prioritize strong personalization for seen classes under extreme data scarcity, but they struggle with weaker generalization to unseen classes under non-IID conditions, which severely restrict their practical utility. To circumvent this problem, we propose a novel personalized federated few-shot learning framework that integrates two levels of complementary learning to simultaneously improve personalization and generalization in such challenging settings. At the intra-client level, a dual-branch architecture is proposed with a global branch capturing holistic image semantics and a local branch focusing on fine-grained features from random crops. These branches interact via knowledge distillation and leverage the synergy between global context and local details to enable robust representation learning from scarce data. At the inter-client level, we propose a collaborative aggregation strategy that prioritizes partners with maximally complementary feature distributions. By integrating knowledge from both similar and diverse clients quantified by maximum mean discrepancy-based prototype discrepancy, our approach effectively leverages cross-federation information. This dual-level strategy ensures strong personalization for seen classes and robust generalization to unseen classes. Experiments on benchmark datasets confirm the effectiveness of our proposed method.
- 30 April 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Multimedia
- Vol. PP (15209210) , 1-11
Modality-Incomplete Multimodal Federated Recommendation with Bidirectional Modality Reconstruction
The aim of federated multimodal recommendation is to leverage rich multimodal information about items to predict user preferences while protecting user privacy. However, existing federated multimodal recommendation algorithms assume that the multimodal information of items held by clients is complete, whereas in real scenarios, such information is often noisy and incomplete. Traditional federated multimodal recommendation algorithms cannot be directly extended to scenarios where modalities are incomplete. To address this issue, this paper proposes a novel federated multimodal recommendation algorithm for modalities-incomplete scenarios. Specifically, to address the issue of missing modalities, we design a multi-scale residual generation network that is pre-trained on a large public dataset to reconstruct the missing modality information. Furthermore, to mitigate the imbalance in how different modalities influence the fused representation across clients, we propose an additive attention module based on noise-conditioned information entropy to dynamically guide modality fusion. Experiments on both modality-complete and modality-incomplete recommendation scenarios demonstrate the superior performance of the proposed algorithm.
- 30 April 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Multimedia
- Vol. PP (15209210) , 1-11
Modality-Incomplete Multimodal Federated Recommendation with Bidirectional Modality Reconstruction
The aim of federated multimodal recommendation is to leverage rich multimodal information about items to predict user preferences while protecting user privacy. However, existing federated multimodal recommendation algorithms assume that the multimodal information of items held by clients is complete, whereas in real scenarios, such information is often noisy and incomplete. Traditional federated multimodal recommendation algorithms cannot be directly extended to scenarios where modalities are incomplete. To address this issue, this paper proposes a novel federated multimodal recommendation algorithm for modalities-incomplete scenarios. Specifically, to address the issue of missing modalities, we design a multi-scale residual generation network that is pre-trained on a large public dataset to reconstruct the missing modality information. Furthermore, to mitigate the imbalance in how different modalities influence the fused representation across clients, we propose an additive attention module based on noise-conditioned information entropy to dynamically guide modality fusion. Experiments on both modality-complete and modality-incomplete recommendation scenarios demonstrate the superior performance of the proposed algorithm.
- 12 February 2026
- Published by Elsevier in Pattern Recognition
- Vol. 177
PF2SMIS: Personalized federated few-shot learning for medical image segmentation
- 12 February 2026
- Published by Elsevier in Pattern Recognition
- Vol. 177
PF2SMIS: Personalized federated few-shot learning for medical image segmentation
- 9 February 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Computational Social Systems
- Vol. 13 (3) , 3655-3667
Towards Adaptive Personalized Federated Meta-Learning Recommendation
Federated recommendation systems aim to provide high-quality recommendations while protecting user privacy. However, existing federated recommendation algorithms typically use a unified item embedding framework, where all clients share the same item representations. Such approaches often fail to capture users’ personalized perceptions of the same item and struggle to reflect subtle changes in user preferences, limiting the effectiveness of personalized recommendations. Moreover, statistical heterogeneity among clients poses additional challenges for model optimization. To address these issues, we propose APFMRec, a personalized federated recommendation framework that enhances both personalization and global optimization. APFMRec incorporates an adaptive item embedding module to dynamically adjust item representations based on individual user preferences. In addition, a meta-learning update module is designed to mitigate statistical heterogeneity and improve collaborative optimization across clients. Extensive experiments on five real-world datasets demonstrate the effectiveness of APFMRec. The proposed framework consistently outperforms existing federated recommendation methods, achieving up to 8.1% improvement in HR@10 and 7.8% improvement in NDCG@10.
- 9 February 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Computational Social Systems
- Vol. 13 (3) , 3655-3667
Towards Adaptive Personalized Federated Meta-Learning Recommendation
Federated recommendation systems aim to provide high-quality recommendations while protecting user privacy. However, existing federated recommendation algorithms typically use a unified item embedding framework, where all clients share the same item representations. Such approaches often fail to capture users’ personalized perceptions of the same item and struggle to reflect subtle changes in user preferences, limiting the effectiveness of personalized recommendations. Moreover, statistical heterogeneity among clients poses additional challenges for model optimization. To address these issues, we propose APFMRec, a personalized federated recommendation framework that enhances both personalization and global optimization. APFMRec incorporates an adaptive item embedding module to dynamically adjust item representations based on individual user preferences. In addition, a meta-learning update module is designed to mitigate statistical heterogeneity and improve collaborative optimization across clients. Extensive experiments on five real-world datasets demonstrate the effectiveness of APFMRec. The proposed framework consistently outperforms existing federated recommendation methods, achieving up to 8.1% improvement in HR@10 and 7.8% improvement in NDCG@10.
- Open Access
- 10 November 2025
- Published by Association for Computing Machinery (ACM)
- p. 686-695
DT-FedSDC: A Dual-Target Federated Framework with Semantic Enhancement and Disentangled Contrastive Learning for Cross-Domain Recommendation
- 10 November 2025
- Published by Association for Computing Machinery (ACM)
- p. 686-695
Open AccessDT-FedSDC: A Dual-Target Federated Framework with Semantic Enhancement and Disentangled Contrastive Learning for Cross-Domain Recommendation
- Open Access
- 11 April 2025
- Published by Association for the Advancement of Artificial Intelligence (AAAI) in Proceedings of the AAAI Conference on Artificial Intelligence
- Vol. 39 (20) , 21243-21251
FedFSL-CFRD: Personalized Federated Few-Shot Learning with Collaborative Feature Representation Disentanglement
Federated few-shot learning (FedFSL) aims to enable the clients to obtain personalized generalization models for unseen categories with only a small number of referenceable samples in the distributed collaborative training paradigm. Most existing FedFSL-related algorithms suffer from domain bias and feature coupling in the presence of data heterogeneity and sample scarcity. In this work, we propose a collaborative feature representation disentanglement (CFRD) scheme for FedFSL to address these issues. After each client receives the global aggregation parameters, the original feature representation is decoupled into global communal features and local personality features with personalized bias representation, to maintain both global consistency and local relevance in the first feature representation disentanglement. On the few-shot metric space about the second feature representation disentanglement, category-independent information is encoded by class-specific and class-irrelevant reconstructions to separate the discriminative features. The proposed scheme collaboratively accomplishes global domain bias feature disentanglement and local category degradation feature disentanglement from client-wise and class-wise. Experiments on three few-shot benchmark datasets conforming to the FedFSL paradigm demonstrate that our proposed method outperforms state-of-the-art approaches in both global generality and local specificity.
- 11 April 2025
- Published by Association for the Advancement of Artificial Intelligence (AAAI) in Proceedings of the AAAI Conference on Artificial Intelligence
- Vol. 39 (20) , 21243-21251
Open AccessFedFSL-CFRD: Personalized Federated Few-Shot Learning with Collaborative Feature Representation Disentanglement
Federated few-shot learning (FedFSL) aims to enable the clients to obtain personalized generalization models for unseen categories with only a small number of referenceable samples in the distributed collaborative training paradigm. Most existing FedFSL-related algorithms suffer from domain bias and feature coupling in the presence of data heterogeneity and sample scarcity. In this work, we propose a collaborative feature representation disentanglement (CFRD) scheme for FedFSL to address these issues. After each client receives the global aggregation parameters, the original feature representation is decoupled into global communal features and local personality features with personalized bias representation, to maintain both global consistency and local relevance in the first feature representation disentanglement. On the few-shot metric space about the second feature representation disentanglement, category-independent information is encoded by class-specific and class-irrelevant reconstructions to separate the discriminative features. The proposed scheme collaboratively accomplishes global domain bias feature disentanglement and local category degradation feature disentanglement from client-wise and class-wise. Experiments on three few-shot benchmark datasets conforming to the FedFSL paradigm demonstrate that our proposed method outperforms state-of-the-art approaches in both global generality and local specificity.
- 14 March 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Emerging Topics in Computational Intelligence
- Vol. 9 (3) , 2359-2370
HGRL-S: Towards Heterogeneous Graph Representation Learning With Optimized Structures
Heterogeneous Graph Neural Networks (HetGNN) have garnered significant attention and demonstrated success in tackling various tasks. However, most existing HetGNNs face challenges in effectively addressing unreliable heterogeneous graph structures and encounter semantic indistinguishability problems as their depth increases. In an effort to deal with these challenges, we introduce a novel heterogeneous graph representation learning with optimized structures to optimize heterogeneous graph structures and utilize semantic aggregation mechanism to alleviate semantic indistinguishability while learning node embeddings. To address the heterogeneity of relations within heterogeneous graphs, the proposed algorithm employs a strategy of generating distinct relational subgraphs and incorporating them with node features to optimize structural learning. To resolve the issue of semantic indistinguishability, the proposed algorithm adopts a semantic aggregation mechanism to assign appropriate weights to different meta-paths, consequently enhancing the effectiveness of captured node features. This methodology enables the learning of distinguishable node embeddings by a deeper HetGNN model. Extensive experiments on the node classification task validate the promising performance of the proposed framework when compared with state-of-the-art methods.
- 14 March 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Emerging Topics in Computational Intelligence
- Vol. 9 (3) , 2359-2370
HGRL-S: Towards Heterogeneous Graph Representation Learning With Optimized Structures
Heterogeneous Graph Neural Networks (HetGNN) have garnered significant attention and demonstrated success in tackling various tasks. However, most existing HetGNNs face challenges in effectively addressing unreliable heterogeneous graph structures and encounter semantic indistinguishability problems as their depth increases. In an effort to deal with these challenges, we introduce a novel heterogeneous graph representation learning with optimized structures to optimize heterogeneous graph structures and utilize semantic aggregation mechanism to alleviate semantic indistinguishability while learning node embeddings. To address the heterogeneity of relations within heterogeneous graphs, the proposed algorithm employs a strategy of generating distinct relational subgraphs and incorporating them with node features to optimize structural learning. To resolve the issue of semantic indistinguishability, the proposed algorithm adopts a semantic aggregation mechanism to assign appropriate weights to different meta-paths, consequently enhancing the effectiveness of captured node features. This methodology enables the learning of distinguishable node embeddings by a deeper HetGNN model. Extensive experiments on the node classification task validate the promising performance of the proposed framework when compared with state-of-the-art methods.
- 31 January 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Geoscience and Remote Sensing
- Vol. 63 (01962892) , 1-12
Toward Federated Customized Neural Architecture Search for Remote Sensing Scene Classification
Remote sensing scenarios usually involve sensitive geographic information on national security and regional development. In the commonly used centralized machine learning paradigm, data dispersed in various locations are concentrated and processed on a single server, which is prone to privacy leakage and data security concerns. Besides, it is difficult to solve the high heterogeneity of remote sensing images by simply applying federated learning algorithms to scene classification. In this paper, we formulate a federated remote sensing scene classification (FedSC) framework, and design a customized neural architecture search (CNAS) to achieve both global generality for multi-party collaborative distributed training and local specificity for personalized remote sensing scene customization. The proposed FedSC is generalizable to be implemented in any mannually designed networks, network pruning strategies, or NAS methods related to remote sensing scene classification. While the designed CNAS not only achieves collaborative distributed training in protecting participant data privacy to obtain a generalized global model, but also provides a customized local model for each participant that is more in line with the characteristics of private remote sensing scenarios. Overall, the proposed FedSCCNAS provides a novel federated collaborative training paradigm for remote sensing scene classification in terms of data privacy, data heterogeneity, and personalized customization. Extensive analytical and comparative experiments on three benchmark remote sensing scene classification datasets validate the versatility and effectiveness of our methods, the proposed FedSCCNAS exhibits superior competitiveness compared to state-of-the-art methods.
- 31 January 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Geoscience and Remote Sensing
- Vol. 63 (01962892) , 1-12
Toward Federated Customized Neural Architecture Search for Remote Sensing Scene Classification
Remote sensing scenarios usually involve sensitive geographic information on national security and regional development. In the commonly used centralized machine learning paradigm, data dispersed in various locations are concentrated and processed on a single server, which is prone to privacy leakage and data security concerns. Besides, it is difficult to solve the high heterogeneity of remote sensing images by simply applying federated learning algorithms to scene classification. In this paper, we formulate a federated remote sensing scene classification (FedSC) framework, and design a customized neural architecture search (CNAS) to achieve both global generality for multi-party collaborative distributed training and local specificity for personalized remote sensing scene customization. The proposed FedSC is generalizable to be implemented in any mannually designed networks, network pruning strategies, or NAS methods related to remote sensing scene classification. While the designed CNAS not only achieves collaborative distributed training in protecting participant data privacy to obtain a generalized global model, but also provides a customized local model for each participant that is more in line with the characteristics of private remote sensing scenarios. Overall, the proposed FedSCCNAS provides a novel federated collaborative training paradigm for remote sensing scene classification in terms of data privacy, data heterogeneity, and personalized customization. Extensive analytical and comparative experiments on three benchmark remote sensing scene classification datasets validate the versatility and effectiveness of our methods, the proposed FedSCCNAS exhibits superior competitiveness compared to state-of-the-art methods.
- 13 January 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Computational Social Systems
- Vol. 12 (5) , 2986-2998
Personalized Federated Contrastive Learning for Recommendation
Recommender systems play crucial roles in addressing the issue of information overload, but traditional centralized storage in recommendation poses significant privacy concerns. In recent years, federated learning has been successfully introduced into a recommendation, while these algorithms still encounter several challenges. First, real-world recommendation scenarios often suffer from sparse data, making it difficult for models to learn reliable representations. Second, data heterogeneity necessitates the design of personalized models to enhance recommendation performance. To address these challenges, we propose a federated recommendation approach based on graph neural networks, named federated personalized contrastive learning for recommendation. On the client side, we propose a contrastive learning approach to enhance the embedding quality of nodes (users or items) by maximizing positive similarities. Specifically, we formulate the concept of structural neighbors based on the graph structure and devise a contrastive learning objective. We treat nodes and their structural neighbors as positive pairs to better learn node representations. On the server side, we group users based on the learned representations and compute cluster-level federated models and a global model. Each user learns a personalized model by combining these two models. Extensive experiments on five real-world datasets demonstrate that the proposed algorithm outperforms existing methods in terms of performance.
- 13 January 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Computational Social Systems
- Vol. 12 (5) , 2986-2998
Personalized Federated Contrastive Learning for Recommendation
Recommender systems play crucial roles in addressing the issue of information overload, but traditional centralized storage in recommendation poses significant privacy concerns. In recent years, federated learning has been successfully introduced into a recommendation, while these algorithms still encounter several challenges. First, real-world recommendation scenarios often suffer from sparse data, making it difficult for models to learn reliable representations. Second, data heterogeneity necessitates the design of personalized models to enhance recommendation performance. To address these challenges, we propose a federated recommendation approach based on graph neural networks, named federated personalized contrastive learning for recommendation. On the client side, we propose a contrastive learning approach to enhance the embedding quality of nodes (users or items) by maximizing positive similarities. Specifically, we formulate the concept of structural neighbors based on the graph structure and devise a contrastive learning objective. We treat nodes and their structural neighbors as positive pairs to better learn node representations. On the server side, we group users based on the learned representations and compute cluster-level federated models and a global model. Each user learns a personalized model by combining these two models. Extensive experiments on five real-world datasets demonstrate that the proposed algorithm outperforms existing methods in terms of performance.
- 13 June 2023
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Neural Networks and Learning Systems
- Vol. 35 (10) , 14932-14944
Device-Performance-Driven Heterogeneous Multiparty Learning for Arbitrary Images
Multiparty learning (MPL) is an emerging framework for privacy-preserving collaborative learning. It enables individual devices to build a knowledge-shared model and remaining sensitive data locally. However, with the continuous increase of users, the heterogeneity gap between data and equipment becomes wider, which leads to the problem of model heterogeneous. In this article, we concentrate on two practical issues: data heterogeneous problem and model heterogeneous problem, and propose a novel personal MPL method named device-performance-driven heterogeneous MPL (HMPL). First, facing the data heterogeneous problem, we focus on the problem of various devices holding arbitrary data sizes. We introduce a heterogeneous feature-map integration method to adaptively unify the various feature maps. Meanwhile, to handle the model heterogeneous problem, as it is essential to customize models for adapting to the various computing performances, we propose a layer-wise model generation and aggregation strategy. The method can generate customized models based on the device’s performance. In the aggregation process, the shared model parameters are updated through the rules that the network layers with the same semantics are aggregated with each other. Extensive experiments are conducted on four popular datasets, and the result demonstrates that our proposed framework outperforms the state of the art (SOTA).
- 13 June 2023
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Neural Networks and Learning Systems
- Vol. 35 (10) , 14932-14944
Device-Performance-Driven Heterogeneous Multiparty Learning for Arbitrary Images
Multiparty learning (MPL) is an emerging framework for privacy-preserving collaborative learning. It enables individual devices to build a knowledge-shared model and remaining sensitive data locally. However, with the continuous increase of users, the heterogeneity gap between data and equipment becomes wider, which leads to the problem of model heterogeneous. In this article, we concentrate on two practical issues: data heterogeneous problem and model heterogeneous problem, and propose a novel personal MPL method named device-performance-driven heterogeneous MPL (HMPL). First, facing the data heterogeneous problem, we focus on the problem of various devices holding arbitrary data sizes. We introduce a heterogeneous feature-map integration method to adaptively unify the various feature maps. Meanwhile, to handle the model heterogeneous problem, as it is essential to customize models for adapting to the various computing performances, we propose a layer-wise model generation and aggregation strategy. The method can generate customized models based on the device’s performance. In the aggregation process, the shared model parameters are updated through the rules that the network layers with the same semantics are aggregated with each other. Extensive experiments are conducted on four popular datasets, and the result demonstrates that our proposed framework outperforms the state of the art (SOTA).