Flexible Differentially Private Vertical Federated Learning with Adaptive Feature Embeddings
Not on the map. Why?
Abstract (arXiv)
The emergence of vertical federated learning (VFL) has stimulated concerns about the imperfection in privacy protection, as shared feature embeddings may reveal sensitive information under privacy attacks. This paper studies the delicate equilibrium between data privacy and task utility goals of VFL under differential privacy (DP). To address the generality issue of prior arts, this paper advocates a flexible and generic approach that decouples the two goals and addresses them successively. Specifically, we initially derive a rigorous privacy guarantee by applying norm clipping on shared feature embeddings, which is applicable across various datasets and models. Subsequently, we demonstrate that task utility can be optimized via adaptive adjustments on the scale and distribution of feature embeddings in an accuracy-appreciative way, without compromising established DP mechanisms. We concretize our observation into the proposed VFL-AFE framework, which exhibits effectiveness against privacy attacks and the capacity to retain favorable task utility, as substantiated by extensive experiments.
Key facts. Cited by 1 paper in the map (data through 14 Oct 2026). Builds on 20 of its 41 references that are in the map. First posted on arXiv 26 Jul 2023.
Builds on
20 of 41 references.
References · 20
- 2020A Simple Framework for Contrastive Learning of Visual Representationscited by 12,049
- 2016Communication-Efficient Learning of Deep Networks from Decentralized Datacited by 7,669
- 2020Supervised Contrastive Learningcited by 2,829
- 2016Membership Inference Attacks against Machine Learning Modelscited by 2,281
- 2021Enhanced Membership Inference Attacks against Machine Learning Modelscited by 248
- 2017Private federated learning on vertically partitioned data via entity resolution and additively homomorphic encryptioncited by 176
- 2020Feature Inference Attack on Model Predictions in Vertical Federated Learningcited by 106
- 2020Privacy Preserving Vertical Federated Learning for Tree-based Modelscited by 89
- 2020VAFL: a Method of Vertical Asynchronous Federated Learningcited by 87
- 2021CAFE: Catastrophic Data Leakage in Vertical Federated Learningcited by 76
- 2020SplitNN-driven Vertical Partitioningcited by 42
- 2019A Quasi-Newton Method Based Vertical Federated Learning Framework for Logistic Regressioncited by 38
8 more
- 2021FedV: Privacy-Preserving Federated Learning over Vertically Partitioned Datacited by 31
- 2020Hybrid Differentially Private Federated Learning on Vertically Partitioned Datacited by 22
- 2022Flexible Vertical Federated Learning with Heterogeneous Partiescited by 20
- 2021Efficient passive membership inference attack in federated learningcited by 16
- 2019Learning Privately over Distributed Features: An ADMM Sharing Approachcited by 14
- 2022Feature Reconstruction Attacks and Countermeasures of DNN training in Vertical Federated Learningcited by 13
- 2022Improving Privacy-Preserving Vertical Federated Learning by Efficient Communication with ADMMcited by 10
- 2022Differentially Private Vertical Federated Learningcited by 4
Built on by
Citing papers per year: 2024: 1.
1 paper.