Where Does This Data Come From? Enhanced Source Inference Attacks in Federated Learning

Chen, Haiyang; Xu, Xiaolong; Zhu, Xiang; Zhou, Xiaokang; Dai, Fei; Gao, Yansong; Chen, Xiao; Wang, Shuo; Hu, Hongsheng

doi:10.24963/ijcai.2025/536

Where Does This Data Come From? Enhanced Source Inference Attacks in Federated Learning

Haiyang Chen, Xiaolong Xu, Xiang Zhu, Xiaokang Zhou, Fei Dai, Yansong Gao, Xiao Chen, Shuo Wang, Hongsheng Hu

Download BibTeX Cite

Proceedings of the Thirty-Fourth International Joint Conference on Artificial Intelligence. Main Track. Pages 4815-4823.

https://doi.org/10.24963/ijcai.2025/536

Keywords

Machine Learning: ML: Trustworthy machine learning
Multidisciplinary Topics and Applications: MTA: Security and privacy
AI Ethics, Trust, Fairness: ETF: Trustworthy AI
Machine Learning: ML: Federated learning

Abstract

Federated learning (FL) enables collaborative model training without exposing raw data, offering a privacy-aware alternative to centralized learning. However, FL remains vulnerable to various privacy attacks that exploit shared model updates, including membership inference, property inference, and gradient inversion. Source inference attacks further threaten FL by identifying which client contributed a specific training sample, posing severe risks to user and institutional privacy. Existing source inference attacks mainly assume passive adversaries and overlook more realistic scenarios where the server actively manipulates the training process. In this paper, we present an enhanced source inference attack that demonstrates how a malicious server can amplify behavioral differences between clients to more accurately infer data origin. Our approach introduces active training manipulation and data augmentation to expose client-specific patterns. Experimental results across five representative FL algorithms and multiple datasets show that our method significantly outperforms prior passive attacks. These findings reveal a deeper level of privacy vulnerability in FL and call for stronger defense mechanisms under active threat models.

Haiyang Chen, Xiaolong Xu, Xiang Zhu, Xiaokang Zhou, Fei Dai, Yansong Gao, Xiao Chen, Shuo Wang, and Hongsheng Hu. 2025. Where Does This Data Come From? Enhanced Source Inference Attacks in Federated Learning. In Proceedings of the Thirty-Fourth International Joint Conference on Artificial Intelligence, IJCAI-25. IJCAI, 4815–4823. https://doi.org/10.24963/ijcai.2025/536

@inproceedings{ijcai2025p536,
  title     = {Where Does This Data Come From? Enhanced Source Inference Attacks in Federated Learning},
  author    = {Chen, Haiyang and Xu, Xiaolong and Zhu, Xiang and Zhou, Xiaokang and Dai, Fei and Gao, Yansong and Chen, Xiao and Wang, Shuo and Hu, Hongsheng},
  booktitle = {Proceedings of the Thirty-Fourth International Joint Conference on
               Artificial Intelligence, {IJCAI-25}},
  publisher = {International Joint Conferences on Artificial Intelligence Organization},
  editor    = {James Kwok},
  pages     = {4815--4823},
  year      = {2025},
  month     = {8},
  note      = {Main Track},
  doi       = {10.24963/ijcai.2025/536},
  url       = {https://doi.org/10.24963/ijcai.2025/536},
}

TY  - CPAPER
TI  - Where Does This Data Come From? Enhanced Source Inference Attacks in Federated Learning
AU  - Chen, Haiyang
AU  - Xu, Xiaolong
AU  - Zhu, Xiang
AU  - Zhou, Xiaokang
AU  - Dai, Fei
AU  - Gao, Yansong
AU  - Chen, Xiao
AU  - Wang, Shuo
AU  - Hu, Hongsheng
T2  - Proceedings of the Thirty-Fourth International Joint Conference on Artificial Intelligence, IJCAI-25
PB  - International Joint Conferences on Artificial Intelligence Organization
PY  - 2025
SP  - 4815
EP  - 4823
DO  - 10.24963/ijcai.2025/536
UR  - https://doi.org/10.24963/ijcai.2025/536
ER  -