Bad-PFL: Exploring Backdoor Attacks against Personalized Federated Learning
Fuente:
arXiv
Salvato in:
| Autori principali: | , , , |
|---|---|
| Natura: | Preprint |
| Pubblicazione: |
2025
|
| Soggetti: | |
| Accesso online: | |
| Tags: |
Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
|
| _version_ | 1866909463824302080 |
|---|---|
| author | Fan, Mingyuan Hu, Zhanyi Wang, Fuyi Chen, Cen |
| author_facet | Fan, Mingyuan Hu, Zhanyi Wang, Fuyi Chen, Cen |
| contents | Data heterogeneity and backdoor attacks rank among the most significant challenges facing federated learning (FL). For data heterogeneity, personalized federated learning (PFL) enables each client to maintain a private personalized model to cater to client-specific knowledge. Meanwhile, vanilla FL has proven vulnerable to backdoor attacks. However, recent advancements in PFL community have demonstrated a potential immunity against such attacks. This paper explores this intersection further, revealing that existing federated backdoor attacks fail in PFL because backdoors about manually designed triggers struggle to survive in personalized models. To tackle this, we design Bad-PFL, which employs features from natural data as our trigger. As long as the model is trained on natural data, it inevitably embeds the backdoor associated with our trigger, ensuring its longevity in personalized models. Moreover, our trigger undergoes mutual reinforcement training with the model, further solidifying the backdoor's durability and enhancing attack effectiveness. The large-scale experiments across three benchmark datasets demonstrate the superior performance of our attack against various PFL methods, even when equipped with state-of-the-art defense mechanisms. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2501_12736 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Bad-PFL: Exploring Backdoor Attacks against Personalized Federated Learning Fan, Mingyuan Hu, Zhanyi Wang, Fuyi Chen, Cen Machine Learning Cryptography and Security Computer Vision and Pattern Recognition Data heterogeneity and backdoor attacks rank among the most significant challenges facing federated learning (FL). For data heterogeneity, personalized federated learning (PFL) enables each client to maintain a private personalized model to cater to client-specific knowledge. Meanwhile, vanilla FL has proven vulnerable to backdoor attacks. However, recent advancements in PFL community have demonstrated a potential immunity against such attacks. This paper explores this intersection further, revealing that existing federated backdoor attacks fail in PFL because backdoors about manually designed triggers struggle to survive in personalized models. To tackle this, we design Bad-PFL, which employs features from natural data as our trigger. As long as the model is trained on natural data, it inevitably embeds the backdoor associated with our trigger, ensuring its longevity in personalized models. Moreover, our trigger undergoes mutual reinforcement training with the model, further solidifying the backdoor's durability and enhancing attack effectiveness. The large-scale experiments across three benchmark datasets demonstrate the superior performance of our attack against various PFL methods, even when equipped with state-of-the-art defense mechanisms. |
| title | Bad-PFL: Exploring Backdoor Attacks against Personalized Federated Learning |
| topic | Machine Learning Cryptography and Security Computer Vision and Pattern Recognition |
| url | https://arxiv.org/abs/2501.12736 |