AdaMixup: A Dynamic Defense Framework for Membership Inference Attack Mitigation

TL;DR

AdaMixup is a novel adaptive mixup-based training framework that enhances deep learning models' privacy against membership inference attacks while preserving high accuracy.

Contribution

It introduces a dynamic, adaptive mixup strategy during training to improve privacy protection against inference attacks.

Findings

Significantly reduces membership inference attack success rate.

Maintains high model accuracy with privacy enhancement.

Demonstrates effectiveness across multiple datasets.

Abstract

Membership inference attacks have emerged as a significant privacy concern in the training of deep learning models, where attackers can infer whether a data point was part of the training set based on the model's outputs. To address this challenge, we propose a novel defense mechanism, AdaMixup. AdaMixup employs adaptive mixup techniques to enhance the model's robustness against membership inference attacks by dynamically adjusting the mixup strategy during training. This method not only improves the model's privacy protection but also maintains high performance. Experimental results across multiple datasets demonstrate that AdaMixup significantly reduces the risk of membership inference attacks while achieving a favorable trade-off between defensive efficiency and model accuracy. This research provides an effective solution for data privacy protection and lays the groundwork for future advancements in mixup training methods.