Distributed deep learning frameworks such as split learning provide great benefits with regards to the computational cost of training deep neural networks and the privacy-aware utilization of the collective data of a group of data-holders. Split learning, in particular, achieves this goal by dividing a neural network between a client and a server so that the client computes the initial set of layers, and the server computes the rest. However, this method introduces a unique attack vector for a malicious server attempting to steal the client's private data: the server can direct the client model towards learning any task of its choice, e.g. towards outputting easily invertible values. With a concrete example already proposed (Pasquini et al., CCS '21), such training-hijacking attacks present a significant risk for the data privacy of split learning clients. In this paper, we propose SplitGuard, a method by which a split learning client can detect whether it is being targeted by a training-hijacking attack or not. We experimentally evaluate our method's effectiveness, compare it with potential alternatives, and discuss in detail various points related to its use. We conclude that SplitGuard can effectively detect training-hijacking attacks while minimizing the amount of information recovered by the adversaries.
翻译:分散的深层次学习框架,例如分解学习,在培训深神经网络的计算成本和对一组数据持有者集体数据的隐私意识利用方面,对于培训深神经网络和一组数据共享群体的隐私利用的保密利用,大有裨益。 特别是,分解学习,通过将客户和服务器分隔一个神经网络,使客户计算初始的一组层,服务器则计算其余部分。 但是,这种方法为恶意服务器试图窃取客户私人数据的恶意服务器引入了一种独特的攻击矢量:服务器可以引导客户模式学习自己选择的任何任务,例如输出容易忽略的值。用一个已经提出的具体例子(Pasquini等人,CC'21),这种劫持培训袭击对分解学习客户的数据隐私构成了重大风险。 在本文中,我们提出SplitGuard,一个分裂学习客户能够检测其是否被培训劫持攻击的目标是否为攻击目标的方法。 我们实验性地评估了我们的方法的有效性,将它与潜在的替代方法进行比较,并详细讨论与攻击相关的各点(Pasquinial Guard) 。 我们得出结论,在尽可能减少攻击数量时,我们通过Slistalgualgualgard检测。