PRIVEE: Privacy-Preserving Vertical Federated Learning Against Feature Inference Attacks
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_ | 1866911318719594496 |
|---|---|
| author | Madabushi, Sindhuja Khan, Ahmad Faraz Ali, Haider Swami, Ananthram Ning, Rui Wu, Hongyi Cho, Jin-Hee |
| author_facet | Madabushi, Sindhuja Khan, Ahmad Faraz Ali, Haider Swami, Ananthram Ning, Rui Wu, Hongyi Cho, Jin-Hee |
| contents | Vertical Federated Learning (VFL) enables collaborative model training across organizations that share common user samples but hold disjoint feature spaces. Despite its potential, VFL is susceptible to feature inference attacks, in which adversarial parties exploit shared confidence scores (i.e., prediction probabilities) during inference to reconstruct private input features of other participants. To counter this threat, we propose PRIVEE (PRIvacy-preserving Vertical fEderated lEarning), a novel defense mechanism named after the French word privée, meaning "private." PRIVEE obfuscates confidence scores while preserving critical properties such as relative ranking and inter-score distances. Rather than exposing raw scores, PRIVEE shares only the transformed representations, mitigating the risk of reconstruction attacks without degrading model prediction accuracy. Extensive experiments show that PRIVEE achieves a threefold improvement in privacy protection compared to state-of-the-art defenses, while preserving full predictive performance against advanced feature inference attacks. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2512_12840 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | PRIVEE: Privacy-Preserving Vertical Federated Learning Against Feature Inference Attacks Madabushi, Sindhuja Khan, Ahmad Faraz Ali, Haider Swami, Ananthram Ning, Rui Wu, Hongyi Cho, Jin-Hee Machine Learning Artificial Intelligence Vertical Federated Learning (VFL) enables collaborative model training across organizations that share common user samples but hold disjoint feature spaces. Despite its potential, VFL is susceptible to feature inference attacks, in which adversarial parties exploit shared confidence scores (i.e., prediction probabilities) during inference to reconstruct private input features of other participants. To counter this threat, we propose PRIVEE (PRIvacy-preserving Vertical fEderated lEarning), a novel defense mechanism named after the French word privée, meaning "private." PRIVEE obfuscates confidence scores while preserving critical properties such as relative ranking and inter-score distances. Rather than exposing raw scores, PRIVEE shares only the transformed representations, mitigating the risk of reconstruction attacks without degrading model prediction accuracy. Extensive experiments show that PRIVEE achieves a threefold improvement in privacy protection compared to state-of-the-art defenses, while preserving full predictive performance against advanced feature inference attacks. |
| title | PRIVEE: Privacy-Preserving Vertical Federated Learning Against Feature Inference Attacks |
| topic | Machine Learning Artificial Intelligence |
| url | https://arxiv.org/abs/2512.12840 |