SPEAR: Exact Gradient Inversion of Batches in Federated Learning

Dimitar Iliev Dimitrov, Maximilian Baader, Mark Niklas Mueller, Martin Vechev

Keywords: Federated Learning, Exact Gradient Inversion, Gradient Leakage, Privacy, Attack

TL;DR: We present the first algorithm for exact gradient inversion on batch sizes $>1$ to reconstruct inputs in the honest-but-curious federated learning setting.

Abstract: Federated learning is a framework for collaborative machine learning where clients only share gradient updates and not their private data with a server. However, it was recently shown that gradient inversion attacks can reconstruct this data from the shared gradients. In the important honest-but-curious setting, existing attacks enable exact reconstruction only for batch size of $b=1$, with larger batches permitting only approximate reconstruction. In this work, we propose SPEAR, *the first algorithm reconstructing whole batches with $b >1$ exactly*. SPEAR combines insights into the explicit low-rank structure of gradients with a sampling-based algorithm. Crucially, we leverage ReLU-induced gradient sparsity to precisely filter out large numbers of incorrect samples, making a final reconstruction step tractable. We provide an efficient GPU implementation for fully connected networks and show that it recovers high-dimensional ImageNet inputs in batches of up to $b \lesssim 25$ exactly while scaling to large networks. Finally, we show theoretically that much larger batches can be reconstructed with high probability given exponential time.

Primary Area: Privacy

Submission Number: 18070