Model Inversion Robustness: Can Transfer Learning Help?
This addresses privacy threats in machine learning for users of sensitive data, offering a novel defense approach that avoids utility degradation seen in prior methods.
The paper tackles the problem of Model Inversion (MI) attacks that reconstruct private training data by proposing TL-DMI, a transfer learning-based defense that limits sensitive information encoding to degrade MI performance, achieving state-of-the-art robustness in experiments.
Model Inversion (MI) attacks aim to reconstruct private training data by abusing access to machine learning models. Contemporary MI attacks have achieved impressive attack performance, posing serious threats to privacy. Meanwhile, all existing MI defense methods rely on regularization that is in direct conflict with the training objective, resulting in noticeable degradation in model utility. In this work, we take a different perspective, and propose a novel and simple Transfer Learning-based Defense against Model Inversion (TL-DMI) to render MI-robust models. Particularly, by leveraging TL, we limit the number of layers encoding sensitive information from private training dataset, thereby degrading the performance of MI attack. We conduct an analysis using Fisher Information to justify our method. Our defense is remarkably simple to implement. Without bells and whistles, we show in extensive experiments that TL-DMI achieves state-of-the-art (SOTA) MI robustness. Our code, pre-trained models, demo and inverted data are available at: https://hosytuyen.github.io/projects/TL-DMI