MARS: A Malignity-Aware Backdoor Defense in Federated Learning
Fuente:
arXiv
Saved in:
| Main Authors: | , , , , , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2025
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866908556991660032 |
|---|---|
| author | Wan, Wei Ning, Yuxuan Huang, Zhicong Hong, Cheng Hu, Shengshan Zhou, Ziqi Zhang, Yechao Zhu, Tianqing Zhou, Wanlei Zhang, Leo Yu |
| author_facet | Wan, Wei Ning, Yuxuan Huang, Zhicong Hong, Cheng Hu, Shengshan Zhou, Ziqi Zhang, Yechao Zhu, Tianqing Zhou, Wanlei Zhang, Leo Yu |
| contents | Federated Learning (FL) is a distributed paradigm aimed at protecting participant data privacy by exchanging model parameters to achieve high-quality model training. However, this distributed nature also makes FL highly vulnerable to backdoor attacks. Notably, the recently proposed state-of-the-art (SOTA) attack, 3DFed (SP2023), uses an indicator mechanism to determine whether the backdoor models have been accepted by the defender and adaptively optimizes backdoor models, rendering existing defenses ineffective. In this paper, we first reveal that the failure of existing defenses lies in the employment of empirical statistical measures that are loosely coupled with backdoor attacks. Motivated by this, we propose a Malignity-Aware backdooR defenSe (MARS) that leverages backdoor energy (BE) to indicate the malicious extent of each neuron. To amplify malignity, we further extract the most prominent BE values from each model to form a concentrated backdoor energy (CBE). Finally, a novel Wasserstein distance-based clustering method is introduced to effectively identify backdoor models. Extensive experiments demonstrate that MARS can defend against SOTA backdoor attacks and significantly outperforms existing defenses. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2509_20383 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | MARS: A Malignity-Aware Backdoor Defense in Federated Learning Wan, Wei Ning, Yuxuan Huang, Zhicong Hong, Cheng Hu, Shengshan Zhou, Ziqi Zhang, Yechao Zhu, Tianqing Zhou, Wanlei Zhang, Leo Yu Cryptography and Security Artificial Intelligence Federated Learning (FL) is a distributed paradigm aimed at protecting participant data privacy by exchanging model parameters to achieve high-quality model training. However, this distributed nature also makes FL highly vulnerable to backdoor attacks. Notably, the recently proposed state-of-the-art (SOTA) attack, 3DFed (SP2023), uses an indicator mechanism to determine whether the backdoor models have been accepted by the defender and adaptively optimizes backdoor models, rendering existing defenses ineffective. In this paper, we first reveal that the failure of existing defenses lies in the employment of empirical statistical measures that are loosely coupled with backdoor attacks. Motivated by this, we propose a Malignity-Aware backdooR defenSe (MARS) that leverages backdoor energy (BE) to indicate the malicious extent of each neuron. To amplify malignity, we further extract the most prominent BE values from each model to form a concentrated backdoor energy (CBE). Finally, a novel Wasserstein distance-based clustering method is introduced to effectively identify backdoor models. Extensive experiments demonstrate that MARS can defend against SOTA backdoor attacks and significantly outperforms existing defenses. |
| title | MARS: A Malignity-Aware Backdoor Defense in Federated Learning |
| topic | Cryptography and Security Artificial Intelligence |
| url | https://arxiv.org/abs/2509.20383 |