When we enforce differential privacy in machine learning, the utility-privacy trade-off is different w.r.t. each group. Gradient clipping and random noise addition disproportionately affect underrepresented and complex classes and subgroups, which results in inequality in utility loss. In this work, we analyze the inequality in utility loss by differential privacy and propose a modified differentially private stochastic gradient descent (DPSGD), called DPSGD-F, to remove the potential disparate impact of differential privacy on the protected group. DPSGD-F adjusts the contribution of samples in a group depending on the group clipping bias such that differential privacy has no disparate impact on group utility. Our experimental evaluation shows how group sample size and group clipping bias affect the impact of differential privacy in DPSGD, and how adaptive clipping for each group helps to mitigate the disparate impact caused by differential privacy in DPSGD-F.
Panjab University（旁遮普大学）；OpenMined（OpenMined组织）；Manipal Institute of Technology（马尼帕尔理工学院）；Carnegie Mellon University（卡内基梅隆大学）；IIT Jodhpur（印度理工学院焦特布尔分校）；Ford Motor Company（福特汽车公司）；Innopolis University（因诺波利斯大学）；University of California, San Diego（加利福尼亚大学圣迭戈分校）；University of Oxford（牛津大学）
Recent advances in differentially private deep learning have demonstrated that application of differential privacy, specifically the DP-SGD algorithm, has a disparate impact on different sub-groups in the population, which leads to a significantly high drop-in model utility for sub-populations that are under-represented (minorities), compared to well-represented ones. In this work, we aim to compare PATE, another mechanism for training deep learning models using differential privacy, with DP-SGD in terms of fairness. We show that PATE does have a disparate impact too, however, it is much less …
Differential Privacy (DP) is an important privacy-enhancing technology for private machine learning systems. It allows to measure and bound the risk associated with an individual participation in a computation. However, it was recently observed that DP learning systems may exacerbate bias and unfairness for different groups of individuals. This paper builds on these important observations and sheds light on the causes of the disparate impacts arising in the problem of differentially private empirical risk minimization. It focuses on the accuracy disparity arising among groups of individuals in…
Mitigating Disparate Impact of Differentially Private Learning through Bounded Adaptive Clipping
Department of Computer Science, University of Helsinki（计算机科学系，赫尔辛基大学）；Department of Electrical and Computer Engineering, American University of Beirut（电气与计算机工程系，贝鲁特美国大学）
Differential privacy (DP) is a popular mechanism for training machine learning models with bounded leakage about the presence of specific points in the training data. The cost of differential privacy is a reduction in the model's accuracy. We demonstrate that in the neural networks trained using differentially private stochastic gradient descent (DP-SGD), this cost is not borne equally: accuracy of DP models drops much more for the underrepresented classes and subgroups. For example, a gender classification model trained using DP-SGD exhibits much lower accuracy for black faces than for white …
As machine learning becomes more widespread throughout society, aspects including data privacy and fairness must be carefully considered, and are crucial for deployment in highly regulated industries. Unfortunately, the application of privacy enhancing technologies can worsen unfair tendencies in models. In particular, one of the most widely used techniques for private model training, differentially private stochastic gradient descent (DPSGD), frequently intensifies disparate impact on groups within data. In this work we study the fine-grained causes of unfairness in DPSGD and identify gradien…
Seldon（塞尔顿）；University of California San Diego（加州大学圣地亚哥分校）；Ford R&A（福特研发中心）；University of Oxford（牛津大学）；OpenMined（开放挖掘）
Deployment of deep learning in different fields and industries is growing day by day due to its performance, which relies on the availability of data and compute. Data is often crowd-sourced and contains sensitive information about its contributors, which leaks into models that are trained on it. To achieve rigorous privacy guarantees, differentially private training mechanisms are used. However, it has recently been shown that differential privacy can exacerbate existing biases in the data and have disparate impacts on the accuracy of different subgroups of data. In this paper, we aim to stud…
Most of privacy protection studies for textual data focus on removing explicit sensitive identifiers. However, personal writing style, as a strong indicator of the authorship, is often neglected. Recent studies, such as SynTF, have shown promising results on privacy-preserving text mining. However, their anonymization algorithm can only output numeric term vectors which are difficult for the recipients to interpret. We propose a novel text generation model with a two-set exponential mechanism for authorship anonymization. By augmenting the semantic information through a REINFORCE training rewa…
Etsy Inc（Etsy公司）；Rutgers University（罗格斯大学）；University at Albany, SUNY（纽约州立大学奥尔巴尼分校）
Recently, there has been an increasing adoption of differential privacy guided algorithms for privacy-preserving machine learning tasks. However, the use of such algorithms comes with trade-offs in terms of algorithmic fairness, which has been widely acknowledged. Specifically, we have empirically observed that the classical collaborative filtering method, trained by differentially private stochastic gradient descent (DP-SGD), results in a disparate impact on user groups with respect to different user engagement levels. This, in turn, causes the original unfair model to become even more biased…
Recent results in coupled or temporal graphical models offer schemes for estimating the relationship structure between features when the data come from related (but distinct) longitudinal sources. A novel application of these ideas is for analyzing group-level differences, i.e., in identifying if trends of estimated objects (e.g., covariance or precision matrices) are different across disparate conditions (e.g., gender or disease). Often, poor effect sizes make detecting the differential signal over the full set of features difficult: for example, dependencies between only a subset of features…
Northeastern University（东北大学）；University of Pennsylvania（宾夕法尼亚大学）
Motivated by settings in which predictive models may be required to be non-discriminatory with respect to certain attributes (such as race), but even collecting the sensitive attribute may be forbidden or restricted, we initiate the study of fair learning under the constraint of differential privacy. We design two learning algorithms that simultaneously promise differential privacy and equalized odds, a 'fairness' condition that corresponds to equalizing false positive and negative rates across protected groups. Our first algorithm is a private implementation of the equalized odds post-process…
University of Edinburgh（爱丁堡大学）；Princeton University（普林斯顿大学）；University of Southern California（南加州大学）
As in traditional machine learning models, models trained with federated learning may exhibit disparate performance across demographic groups. Model holders must identify these disparities to mitigate undue harm to the groups. However, measuring a model's performance in a group requires access to information about group membership which, for privacy reasons, often has limited availability. We propose novel locally differentially private mechanisms to measure differences in performance across groups while protecting the privacy of group membership. To analyze the effectiveness of the mechanisms…
The Alan Turing Institute（艾伦·图灵研究所）；UCLouvain（鲁汶大学）；Università di Genova（热那亚大学）；Imperial College London（帝国理工学院）
Differential privacy is the standard privacy definition for performing analyses over sensitive data. Yet, its privacy budget bounds the number of tasks an analyst can perform with reasonable accuracy, which makes it challenging to deploy in practice. This can be alleviated by private sketching, where the dataset is compressed into a single noisy sketch vector which can be shared with the analysts and used to perform arbitrarily many analyses. However, the algorithms to perform specific tasks from sketches must be developed on a case-by-case basis, which is a major impediment to their use. In t…
Near-Optimal Private Linear Regression via Iterative Hessian Mixing
Department of EECS, Massachusetts Institute of Technology, US（麻省理工学院电子工程与计算机科学系）；School of Computer Science and Engineering, The Hebrew University of Jerusalem, IL（耶路撒冷希伯来大学计算机科学与工程学院）
