arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 1712.09136cs.CR
收藏

关于衡量会员隐私

Towards Measuring Membership Privacy

Yunhui Long, Vincent Bindschaedler, Carl A. Gunter

摘要:

机器学习模型越来越多地通过公共查询接口提供给大众。最近的学术工作已经证明,可以查询这种模型的恶意用户能够推断出关于训练数据内的记录的敏感信息。差分隐私可以阻止这种攻击,但并不是所有的模型都可以很容易地训练来实现这种保证,或者以可接受的效用损失来实现它。因此,如果一个模型是在没有差分隐私保证的情况下训练的,那么我们对释放它的隐私风险知之甚少或可以说。在这项工作中,我们调查和分析了成员攻击,以了解它们为什么以及如何成功。基于这种理解,我们提出了差分训练隐私(DTP),一个经验性的度量,以估计隐私风险发布分类时,如差分隐私的方法不能应用。DTP是一个衡量分类器相对于其训练数据集,我们表明,计算DTP是有效的,在许多实际情况下。我们使用最先进的机器学习模型(例如在真实世界数据集上训练的神经网络)来经验性地验证DTP。我们的研究结果表明,DTP是高度预测的成功成员攻击,因此减少DTP也降低了隐私风险。我们提倡在考虑发布分类器时将DTP用作决策过程的一部分。为此,我们还建议采用DTP-1假设:如果分类器的DTP值大于1,则不应发布。

英文摘要:

Machine learning models are increasingly made available to the masses through public query interfaces. Recent academic work has demonstrated that malicious users who can query such models are able to infer sensitive information about records within the training data. Differential privacy can thwart such attacks, but not all models can be readily trained to achieve this guarantee or to achieve it with acceptable utility loss. As a result, if a model is trained without differential privacy guarantee, little is known or can be said about the privacy risk of releasing it. In this work, we investigate and analyze membership attacks to understand why and how they succeed. Based on this understanding, we propose Differential Training Privacy (DTP), an empirical metric to estimate the privacy risk of publishing a classier when methods such as differential privacy cannot be applied. DTP is a measure of a classier with respect to its training dataset, and we show that calculating DTP is efficient in many practical cases. We empirically validate DTP using state-of-the-art machine learning models such as neural networks trained on real-world datasets. Our results show that DTP is highly predictive of the success of membership attacks and therefore reducing DTP also reduces the privacy risk. We advocate for DTP to be used as part of the decision-making process when considering publishing a classifier. To this end, we also suggest adopting the DTP-1 hypothesis: if a classifier has a DTP value above 1, it should not be published.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑