Knock Knock, Who's There? Membership Inference on Aggregate Location Data

Abstract

Aggregate location data is often used to support smart services and applications, e.g., generating live traffic maps or predicting visits to businesses. In this paper, we present the first study on the feasibility of membership inference attacks on aggregate location time-series. We introduce a game-based definition of the adversarial task, and cast it as a classification problem where machine learning can be used to distinguish whether or not a target user is part of the aggregates.We empirically evaluate the power of these attacks on both raw and differentially private aggregates using two mobility datasets. We find that membership inference is a serious privacy threat, and show how its effectiveness depends on the adversary's prior knowledge, the characteristics of the underlying location data, as well as the number of users and the timeframe on which aggregation is performed. Although differentially private mechanisms can indeed reduce the extent of the attacks, they also yield a significant loss in utility. Moreover, a strategic adversary mimicking the behavior of the defense mechanism can greatly limit the protection they provide. Overall, our work presents a novel methodology geared to evaluate membership inference on aggregate location data in real-world settings and can be used by providers to assess the quality of privacy protection before data release or by regulators to detect violations.

Demystifying Membership Inference Attacks in Machine Learning as a Service

“…Even with a modest privacy parameter of = 8 the authors report an attack accuracy 58.3% with a training accuracy of just 68.6%. The authors in [46] also remark that defense mechanisms based on differential privacy are not always effective, particularly when an attacker is able to mimic the behavior of the perturbation.…”

Auditing Data Provenance in Text-Generation Models

“…Membership inference. Membership inference attacks involve observing the output of some computations over a hidden dataset D and determining whether a specific data point is a member of D. Membership inference attacks against aggregate statistics have been demonstrated in the context of genomic studies [13], location time-series [26], and noisy statistics in general [8].…”

Membership Inference Attacks on Sequence-to-Sequence Models: Is My Data In Your Machine Translation System?

“…Nasr et al (2019) extends the analysis to white-box attacks and a federated learning setting. Pyrgelis et al (2018) provides an empirical study on location data. Veale et al (2018) discusses membership inference and the related model inversion problem, in the context of data protection laws like GDPR.…”