In federated learning (FL), data does not leave personal devices when they are jointly training a machine learning model. Instead, these devices share gradients, parameters, or other model updates, with a central party (e.g., a company) coordinating the training. Because data never "leaves" personal devices, FL is often presented as privacy-preserving. Yet, recently it was shown that this protection is but a thin facade, as even a passive, honest-but-curious attacker observing gradients can reconstruct data of individual users contributing to the protocol. In this work, we show a novel data reconstruction attack which allows an active and dishonest central party to efficiently extract user data from the received gradients. While prior work on data reconstruction in FL relies on solving computationally expensive optimization problems or on making easily detectable modifications to the shared model's architecture or parameters, in our attack the central party makes inconspicuous changes to the shared model's weights before sending them out to the users. We call the modified weights of our attack trap weights. Our active attacker is able to recover user data perfectly, i.e., with zero error, even when this data stems from the same class. Recovery comes with near-zero costs: the attack requires no complex optimization objectives. Instead, our attacker exploits inherent data leakage from model gradients and simply amplifies this effect by maliciously altering the weights of the shared model through the trap weights. These specificities enable our attack to scale to fully-connected and convolutional deep neural networks trained with large mini-batches of data. For example, for the high-dimensional vision dataset ImageNet, we perfectly reconstruct more than 50% of the training data points from mini-batches as large as 100 data points.
翻译:在联邦学习中,多个设备共同训练机器学习模型时,数据并不离开设备。相反,这些设备共享梯度、参数或其他模型更新,并由一个中央方(如公司)协调训练过程。由于数据从未“离开”设备,因此联邦学习通常被认为具有隐私保护功能。然而,最近的研究表明,这种保护只是一个薄弱的幌子。即使是一个被动、诚实但好奇的攻击者观察梯度,也能够重构参与协议的个人用户的数据。在本文中,我们展示了一种新型的数据重构攻击,允许一个活跃且不诚实的中央方从接收到的梯度中有效地提取用户数据。虽然联邦学习中的数据重构之前的工作依赖于求解计算复杂度昂贵的优化问题或对共享模型的架构或参数进行容易检测的修改,但在我们的攻击中,中央方在发送共享模型的权重之前对其进行了不引人注目的修改。我们称攻击中修改后的共享模型的权重为陷阱权重。我们的主动攻击者能够完美地恢复用户数据,即使这些数据来自相同的类别也是如此,并且几乎没有成本:攻击不需要复杂的优化目标。相反,我们的攻击者利用模型梯度的固有数据泄漏,并通过陷阱权重恶意地改变共享模型的权重来放大这种影响。这些特定性使我们的攻击可以扩展到使用大型小批量数据进行训练的全连接和卷积深度神经网络。例如,在高维视觉数据集ImageNet中,我们可以完美地从100个数据点的小批量中重构超过50%的训练数据点。