Plug & Play Attacks: Towards Robust and Flexible Model Inversion Attacks

Lukas Struppek · Dominik Hintersdorf · Antonio De Almeida Correia · Antonia Adler · Kristian Kersting

Hall E #920

Keywords: [ DL: Generative Models and Autoencoders ] [ SA: Privacy-preserving Statistics and Machine Learning ] [ SA: Fairness, Equity, Justice and Safety ] [ SA: Trustworthy Machine Learning ]


Model inversion attacks (MIAs) aim to create synthetic images that reflect the class-wise characteristics from a target classifier's private training data by exploiting the model's learned knowledge. Previous research has developed generative MIAs that use generative adversarial networks (GANs) as image priors tailored to a specific target model. This makes the attacks time- and resource-consuming, inflexible, and susceptible to distributional shifts between datasets. To overcome these drawbacks, we present Plug & Play Attacks, which relax the dependency between the target model and image prior, and enable the use of a single GAN to attack a wide range of targets, requiring only minor adjustments to the attack. Moreover, we show that powerful MIAs are possible even with publicly available pre-trained GANs and under strong distributional shifts, for which previous approaches fail to produce meaningful results. Our extensive evaluation confirms the improved robustness and flexibility of Plug & Play Attacks and their ability to create high-quality images revealing sensitive class characteristics.

Chat is not available.