FedPass: Privacy-Preserving Vertical Federated Deep Learning with Adaptive Obfuscation

TL;DR

FedPass is a novel framework for vertical federated deep learning that uses adaptive obfuscation to protect private features and labels, balancing privacy and model accuracy effectively.

Contribution

It introduces a general privacy-preserving VFL framework with adaptive obfuscation and provides theoretical privacy guarantees.

Findings

Proves strong privacy preservation for features and labels.

Demonstrates superior privacy-performance trade-off in experiments.

Outperforms existing methods across multiple datasets and architectures.

Abstract

Vertical federated learning (VFL) allows an active party with labeled feature to leverage auxiliary features from the passive parties to improve model performance. Concerns about the private feature and label leakage in both the training and inference phases of VFL have drawn wide research attention. In this paper, we propose a general privacy-preserving vertical federated deep learning framework called FedPass, which leverages adaptive obfuscation to protect the feature and label simultaneously. Strong privacy-preserving capabilities about private features and labels are theoretically proved (in Theorems 1 and 2). Extensive experimental result s with different datasets and network architectures also justify the superiority of FedPass against existing methods in light of its near-optimal trade-off between privacy and model performance.