VFLIP: A Backdoor Defense for Vertical Federated Learning via Identification and Purification
Fuente:
arXiv
Saved in:
| Main Authors: | , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2024
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866909299513491456 |
|---|---|
| author | Cho, Yungi Han, Woorim Yu, Miseon Lee, Younghan Bae, Ho Paek, Yunheung |
| author_facet | Cho, Yungi Han, Woorim Yu, Miseon Lee, Younghan Bae, Ho Paek, Yunheung |
| contents | Vertical Federated Learning (VFL) focuses on handling vertically partitioned data over FL participants. Recent studies have discovered a significant vulnerability in VFL to backdoor attacks which specifically target the distinct characteristics of VFL. Therefore, these attacks may neutralize existing defense mechanisms designed primarily for Horizontal Federated Learning (HFL) and deep neural networks. In this paper, we present the first backdoor defense, called VFLIP, specialized for VFL. VFLIP employs the identification and purification techniques that operate at the inference stage, consequently improving the robustness against backdoor attacks to a great extent. VFLIP first identifies backdoor-triggered embeddings by adopting a participant-wise anomaly detection approach. Subsequently, VFLIP conducts purification which removes the embeddings identified as malicious and reconstructs all the embeddings based on the remaining embeddings. We conduct extensive experiments on CIFAR10, CINIC10, Imagenette, NUS-WIDE, and BankMarketing to demonstrate that VFLIP can effectively mitigate backdoor attacks in VFL. https://github.com/blingcho/VFLIP-esorics24 |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2408_15591 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | VFLIP: A Backdoor Defense for Vertical Federated Learning via Identification and Purification Cho, Yungi Han, Woorim Yu, Miseon Lee, Younghan Bae, Ho Paek, Yunheung Machine Learning Vertical Federated Learning (VFL) focuses on handling vertically partitioned data over FL participants. Recent studies have discovered a significant vulnerability in VFL to backdoor attacks which specifically target the distinct characteristics of VFL. Therefore, these attacks may neutralize existing defense mechanisms designed primarily for Horizontal Federated Learning (HFL) and deep neural networks. In this paper, we present the first backdoor defense, called VFLIP, specialized for VFL. VFLIP employs the identification and purification techniques that operate at the inference stage, consequently improving the robustness against backdoor attacks to a great extent. VFLIP first identifies backdoor-triggered embeddings by adopting a participant-wise anomaly detection approach. Subsequently, VFLIP conducts purification which removes the embeddings identified as malicious and reconstructs all the embeddings based on the remaining embeddings. We conduct extensive experiments on CIFAR10, CINIC10, Imagenette, NUS-WIDE, and BankMarketing to demonstrate that VFLIP can effectively mitigate backdoor attacks in VFL. https://github.com/blingcho/VFLIP-esorics24 |
| title | VFLIP: A Backdoor Defense for Vertical Federated Learning via Identification and Purification |
| topic | Machine Learning |
| url | https://arxiv.org/abs/2408.15591 |