Reconstructing Training Data with Informed Adversaries

Borja Balle, Giovanni Cherubin, Jamie Hayes

Keywords: Reconstruction attacks

Abstract: Given access to a machine learning model, can an adversary reconstruct the model’s training data? This work proposes a formal threat model to study this question, shows that reconstruction attacks are feasible in theory and in practice, and presents preliminary results assessing how different factors of standard machine learning pipelines affect the success of reconstruction. Finally, we empirically evaluate what levels of differential privacy suffice to prevent reconstruction attacks.

Paper Under Submission: The paper is NOT under submission at NeurIPS

Community Implementations: [ 1 code implementation](https://www.catalyzex.com/paper/arxiv:2201.04845/code)