Federated learning (FL) emerged as a promising learning paradigm to enable a multitude of participants to construct a joint ML model without exposing their private training data. Existing FL designs have been shown to exhibit vulnerabilities which can be exploited by adversaries both within and outside of the system to compromise data privacy. However, most current works conduct attacks by leveraging gradients on a small batch of data, which is less practical in FL. In this work, we consider a more practical and interesting scenario in which participants share their epoch-averaged gradients (share gradients after at least 1 epoch of local training) rather than per-example or small batch-averaged gradients as in previous works. We perform the first systematic evaluation of attribute reconstruction attack (ARA) launched by the malicious server in the FL system, and empirically demonstrate that the shared epoch-averaged local model gradients can reveal sensitive attributes of local training data of any victim participant. To achieve this goal, we develop a more effective and efficient gradient matching based method called cos-matching to reconstruct the training data attributes. We evaluate our attacks on a variety of real-world datasets, scenarios, assumptions. Our experiments show that our proposed method achieves better attribute attack performance than most existing baselines.
翻译:联邦学习(FL)是一个大有希望的学习范例,它使许多参与者能够在不披露其私人培训数据的情况下建立联合 ML模型,成为了有希望的学习范例。现有的 FL设计显示,存在一些弱点,这些弱点可以被系统内外的对手利用,以损害数据隐私。然而,大多数目前的工作是通过利用少量数据中的梯度来利用梯度进行攻击,而FL则不那么实用。在这项工作中,我们考虑一种更加实际和有趣的情景,即参与者可以分享其最差的差值梯度(至少经过1个地方培训之后的差度梯度),而不是像以前的工作那样,每个例或小批量平均梯度。我们对FL系统中恶意服务器发起的属性重建攻击(ARA)进行第一次系统评价,并实证地表明,共同的偏差当地模型梯度可以揭示任何受害者参与者当地培训数据的敏感属性。为实现这一目标,我们开发一种更有效和高效的梯度匹配方法,其基础是连接,以重建培训数据属性。我们评估了各种真实世界性攻击基线,而不是我们提出的业绩假设。