Increasing use of machine learning (ML) technologies in privacy-sensitive domains such as medical diagnoses, lifestyle predictions, and business decisions highlights the need to better understand if these ML technologies are introducing leakage of sensitive and proprietary training data. In this paper, we focus on model inversion attacks where the adversary knows non-sensitive attributes about records in the training data and aims to infer the value of a sensitive attribute unknown to the adversary, using only black-box access to the target classification model. We first devise a novel confidence score-based model inversion attribute inference attack that significantly outperforms the state-of-the-art. We then introduce a label-only model inversion attack that relies only on the model's predicted labels but still matches our confidence score-based attack in terms of attack effectiveness. We also extend our attacks to the scenario where some of the other (non-sensitive) attributes of a target record are unknown to the adversary. We evaluate our attacks on two types of machine learning models, decision tree and deep neural network, trained on three real datasets. Moreover, we empirically demonstrate the disparate vulnerability of model inversion attacks, i.e., specific groups in the training dataset (grouped by gender, race, etc.) could be more vulnerable to model inversion attacks.
University of California Davis（加州大学戴维斯分校）；University of Kentucky（肯塔基大学）
To protect sensitive data in training a Generative Adversarial Network (GAN), the standard approach is to use differentially private (DP) stochastic gradient descent method in which controlled noise is added to the gradients. The quality of the output synthetic samples can be adversely affected and the training of the network may not even converge in the presence of these noises. We propose Differentially Private Model Inversion (DPMI) method where the private data is first mapped to the latent space via a public generator, followed by a lower-dimensional DP-GAN with better convergent properti…
Queen Mary University of London（伦敦玛丽女王大学）；Centre for Intelligent Sensing（智能感知中心）
Deep neural networks are increasingly deployed for scene analytics, including to evaluate the attention and reaction of people exposed to out-of-home advertisements. However, the features extracted by a deep neural network that was trained to predict a specific, consensual attribute (e.g. emotion) may also encode and thus reveal information about private, protected attributes (e.g. age or gender). In this work, we focus on such leakage of private information at inference time. We consider an adversary with access to the features extracted by the layers of a deployed neural network and use thes…
Increasing use of ML technologies in privacy-sensitive domains such as medical diagnoses, lifestyle predictions, and business decisions highlights the need to better understand if these ML technologies are introducing leakages of sensitive and proprietary training data. In this paper, we focus on one kind of model inversion attacks, where the adversary knows non-sensitive attributes about instances in the training data and aims to infer the value of a sensitive attribute unknown to the adversary, using oracle access to the target classification model. We devise two novel model inversion attrib…
University of Toronto（多伦多大学）；Vector Institute（向量研究所）；Simon Fraser University（西蒙弗雷泽大学）
Given the ubiquity of deep neural networks, it is important that these models do not reveal information about sensitive data that they have been trained on. In model inversion attacks, a malicious user attempts to recover the private dataset used to train a supervised neural network. A successful model inversion attack should generate realistic and diverse samples that accurately describe each of the classes in the private dataset. In this work, we provide a probabilistic interpretation of model inversion attacks, and formulate a variational objective that accounts for both diversity and accur…
University of Illinois at Urbana-Champaign（伊利诺伊大学厄巴纳-香槟分校）；Shanghai Jiao Tong University（上海交通大学）；Macquarie University（麦考瑞大学）；Data61-CSIRO（澳大利亚联邦科学与工业研究组织Data61研究所）
Deep neural networks (DNNs) have recently been widely adopted in various applications, and such success is largely due to a combination of algorithmic breakthroughs, computation resource improvements, and access to a large amount of data. However, the large-scale data collections required for deep learning often contain sensitive information, therefore raising many privacy concerns. Prior research has shown several successful attacks in inferring sensitive training data information, such as model inversion, membership inference, and generative adversarial networks (GAN) based leakage attacks a…
Recommendation is one of the critical applications that helps users find information relevant to their interests. However, a malicious attacker can infer users' private information via recommendations. Prior work obfuscates user-item data before sharing it with recommendation system. This approach does not explicitly address the quality of recommendation while performing data obfuscation. Moreover, it cannot protect users against private-attribute inference attacks based on recommendations. This work is the first attempt to build a Recommendation with Attribute Protection (RAP) model which sim…
Differentially Private (DP) data release is a promising technique to disseminate data without compromising the privacy of data subjects. However the majority of prior work has focused on scenarios where a single party owns all the data. In this paper we focus on the multi-party setting, where different stakeholders own disjoint sets of attributes belonging to the same group of data subjects. Within the context of linear regression that allow all parties to train models on the complete data without the ability to infer private attributes or identities of individuals, we start with directly appl…
In this paper, we tackle the problem of constructing a differentially private synopsis for the classification analyses. Several the state-of-the-art methods follow the structure of existing classification algorithms and are all iterative, which is suboptimal due to the locally optimal choices and the over-divided privacy budget among many sequentially composed steps. Instead, we propose a new approach, PrivPfC, a new differentially private method for releasing data for classification. The key idea is to privately select an optimal partition of the underlying dataset using the given privacy bud…
University of Wisconsin-Madison（威斯康星大学麦迪逊分校）；Carnegie Mellon University（卡内基梅隆大学）；Microsoft Research（微软研究院）
Private regression has received attention from both database and security communities. Recent work by Fredrikson et al. (USENIX Security 2014) analyzed the functional mechanism (Zhang et al. VLDB 2012) for training linear regression models over medical data. Unfortunately, they found that model accuracy is already unacceptable with differential privacy when $\varepsilon = 5$. We address this issue, presenting an explicit connection between differential privacy and stable learning theory through which a substantially better privacy/utility tradeoff can be obtained. Perhaps more importantly, our…
Breaching FedMD: Image Recovery via Paired-Logits Inversion Attack
The University of Tokyo（东京大学）；Institute for AI Industry Research, Tsinghua University（清华大学人工智能产业研究院）；Shanghai Artificial Intelligence Laboratory（上海人工智能实验室）
本文发现 FedMD 范式存在隐私泄露风险，提出 PLI 攻击，恶意服务器通过训练反演神经网络利用模型间的置信度差距，仅用配对 logits 即可高成功率重建私有图像。
Data privacy is an increasingly important aspect of many real-world Data sources that contain sensitive information may have immense potential which could be unlocked using the right privacy enhancing transformations, but current methods often fail to produce convincing output. Furthermore, finding the right balance between privacy and utility is often a tricky trade-off. In this work, we propose a novel approach for data privatization, which involves two steps: in the first step, it removes the sensitive information, and in the second step, it replaces this information with an independent ran…
Online services routinely mine user data to predict user preferences, make recommendations, and place targeted ads. Recent research has demonstrated that several private user attributes (such as political affiliation, sexual orientation, and gender) can be inferred from such data. Can a privacy-conscious user benefit from personalization while simultaneously protecting her private attributes? We study this question in the context of a rating prediction service based on matrix factorization. We construct a protocol of interactions between the service and users that has remarkable optimality pro…
Syracuse University（雪城大学）；Georgia Institute of Technology（佐治亚理工学院）
A critical concern in data-driven decision making is to build models whose outcomes do not discriminate against some demographic groups, including gender, ethnicity, or age. To ensure non-discrimination in learning tasks, knowledge of the sensitive attributes is essential, while, in practice, these attributes may not be available due to legal and ethical requirements. To address this challenge, this paper studies a model that protects the privacy of the individuals sensitive information while also allowing it to learn non-discriminatory predictors. The method relies on the notion of differenti…
The Alan Turing Institute（阿兰·图灵研究所）；Vector Institute（向量研究所）；Université de Lille（里尔大学）；Inria（法国国家信息与自动化研究所）；CNRS（法国国家科学研究中心）；Centrale Lille（里尔中央理工学院）；Université de Lorraine（洛林大学）；University of Toronto（多伦多大学）
Sharing real-world speech utterances is key to the training and deployment of voice-based services. However, it also raises privacy risks as speech contains a wealth of personal data. Speaker anonymization aims to remove speaker information from a speech utterance while leaving its linguistic and prosodic attributes intact. State-of-the-art techniques operate by disentangling the speaker information (represented via a speaker embedding) from these attributes and re-synthesizing speech based on the speaker embedding of another speaker. Prior research in the privacy community has shown that anon…
Exploring User-level Gradient Inversion with a Diffusion Prior
Vanderbilt University（范德堡大学）；University of Wisconsin-Madison（威斯康星大学麦迪逊分校）；Mitsubishi Electric Research Laboratories（三菱电机研发实验室）；Vanderbilt University Medical Center（范德堡大学医学中心）
A Virtual Private Network (VPN) helps to mitigate security and privacy risks of data transmitting on unsecured network such as public Wi-Fi. However, despite awareness of public Wi-Fi risks becoming increasingly common, the use of VPN when using public Wi-Fi is low. To increase adoption, understanding factors driving user decision to adopt a VPN app is an important first step. This study is the first to achieve this objective using discrete choice experiments (DCEs) to elicit individual preferences of specific attributes of a VPN app. The experiments were run in the United Kingdom (UK) and Jap…
A New Federated Learning Framework Against Gradient Inversion Attacks
The University of Hong Kong（香港大学）；Beijing University of Technology（北京工业大学）；Harbin Institute of Technology, Shenzhen（哈尔滨工业大学（深圳））；UC Santa Cruz（加州大学圣克鲁兹分校）
The rapid developments of mobile devices and online social networks have resulted in increasing attention to Mobile Social Networking (MSN). The explosive growth of mobile-connected and location-aware devices makes it possible and meaningful to do the Proximity-based Mobile Social Networks (PMSNs). Users can discover and make new social interactions easily with physical-proximate mobile users through WiFi/Bluetooth interfaces embedded in their smartphones. However, users enjoy these conveniences at the cost of their growing privacy concerns. To address this problem, we propose a suit of priori…
University of Houston（休斯顿大学）；Guilin University of Electronic Technology（桂林电子科技大学）；Guangdong University of Technology（广东工业大学）
Machine learning is increasingly becoming a powerful tool to make decisions in a wide variety of applications, such as medical diagnosis and autonomous driving. Privacy concerns related to the training data and unfair behaviors of some decisions with regard to certain attributes (e.g., sex, race) are becoming more critical. Thus, constructing a fair machine learning model while simultaneously providing privacy protection becomes a challenging problem. In this paper, we focus on the design of classification model with fairness and differential privacy guarantees by jointly combining functional …
We describe a threat model under which a split network-based federated learning system is susceptible to a model inversion attack by a malicious computational server. We demonstrate that the attack can be successfully performed with limited knowledge of the data distribution by the attacker. We propose a simple additive noise method to defend against model inversion, finding that the method can significantly reduce attack efficacy at an acceptable accuracy trade-off on MNIST. Furthermore, we show that NoPeekNN, an existing defensive method, protects different information from exposure, suggest…
Gradient inversion attack (or input recovery from gradient) is an emerging threat to the security and privacy preservation of Federated learning, whereby malicious eavesdroppers or participants in the protocol can recover (partially) the clients' private data. This paper evaluates existing attacks and defenses. We find that some attacks make strong assumptions about the setup. Relaxing such assumptions can substantially weaken these attacks. We then evaluate the benefits of three proposed defense mechanisms against gradient inversion attacks. We show the trade-offs of privacy leakage and data …
University of Florida（佛罗里达大学）；Washington University in St. Louis（圣路易斯华盛顿大学）
We present a framework to learn privacy-preserving encodings of images that inhibit inference of chosen private attributes, while allowing recovery of other desirable information. Rather than simply inhibiting a given fixed pre-trained estimator, our goal is that an estimator be unable to learn to accurately predict the private attributes even with knowledge of the encoding function. We use a natural adversarial optimization-based formulation for this---training the encoding function against a classifier for the private attribute, with both modeled as deep neural networks. The key contribution…
We propose a generic mechanism to efficiently release differentially private synthetic versions of high-dimensional datasets with high utility. The core technique in our mechanism is the use of copulas. Specifically, we use the Gaussian copula to define dependencies of attributes in the input dataset, whose rows are modelled as samples from an unknown multivariate distribution, and then sample synthetic records through this copula. Despite the inherently numerical nature of Gaussian correlations we construct a method that is applicable to both numerical and categorical attributes alike. Our me…
Zhejiang University（浙江大学）；ZJU-Hangzhou Global Scientific and Technological Innovation Center, Zhejiang University（浙江大学杭州国际科创中心）；Key Laboratory of Blockchain and Cyberspace Governance of Zhejiang Province（浙江省区块链与网络空间治理重点实验室）；Jiaxing Research Institute, Zhejiang University（浙江大学嘉兴研究院）；Zhengzhou Xinda Institute of Advanced Technology（郑州信大先进技术研究院）；National University of Singapore（新加坡国立大学）
Neural networks are susceptible to data inference attacks such as the membership inference attack, the adversarial model inversion attack and the attribute inference attack, where the attacker could infer useful information such as the membership, the reconstruction or the sensitive attributes of a data sample from the confidence scores predicted by the target classifier. In this paper, we propose a method, namely PURIFIER, to defend against membership inference attacks. It transforms the confidence score vectors predicted by the target classifier and makes purified confidence scores indisting…
Hong Kong University of Science and Technology（香港科技大学）；WeBank（微众银行）
Social chatbots, also known as chit-chat chatbots, evolve rapidly with large pretrained language models. Despite the huge progress, privacy concerns have arisen recently: training data of large language models can be extracted via model inversion attacks. On the other hand, the datasets used for training chatbots contain many private conversations between two individuals. In this work, we further investigate the privacy leakage of the hidden states of chatbots trained by language modeling which has not been well studied yet. We show that speakers' personas can be inferred through a simple neur…
Random Erasing vs. Model Inversion: A Promising Defense or a False Hope?
Temasek Laboratories, Singapore University of Technology and Design（Temasek实验室，新加坡技术与设计大学）；The Queen’s University Belfast（女王大学贝尔法斯特分校）；Aarhus University（阿arhus大学）；Nanyang Technological University (NTU)（南洋理工大学（NTU）；河内 Vin 大学）；VinUniversity, Hanoi, Vietnam
Split learning is a distributed training framework that allows multiple parties to jointly train a machine learning model over vertically partitioned data (partitioned by attributes). The idea is that only intermediate computation results, rather than private features and labels, are shared between parties so that raw training data remains private. Nevertheless, recent works showed that the plaintext implementation of split learning suffers from severe privacy risks that a semi-honest adversary can easily reconstruct labels. In this work, we propose \textsf{TPSL} (Transcript Private Split Lear…
Anonymized data is highly valuable to both businesses and researchers. A large body of research has however shown the strong limits of the de-identification release-and-forget model, where data is anonymized and shared. This has led to the development of privacy-preserving query-based systems. Based on the idea of "sticky noise", Diffix has been recently proposed as a novel query-based mechanism satisfying alone the EU Article~29 Working Party's definition of anonymization. According to its authors, Diffix adds less noise to answers than solutions based on differential privacy while allowing f…
Machine learning models often pose a threat to the privacy of individuals whose data is part of the training set. Several recent attacks have been able to infer sensitive information from trained models, including model inversion or attribute inference attacks. These attacks are able to reveal the values of certain sensitive features of individuals who participated in training the model. It has also been shown that several factors can contribute to an increased risk of model inversion, including feature influence. We observe that not all features necessarily share the same level of privacy or …
Georgia Institute of Technology（佐治亚理工学院）；Meta AI；Pennsylvania State University（宾夕法尼亚州立大学）；Virginia Tech（弗吉尼亚理工大学）；Cornell University（康奈尔大学）
Federated learning (FL) aims to perform privacy-preserving machine learning on distributed data held by multiple data owners. To this end, FL requires the data owners to perform training locally and share the gradient updates (instead of the private inputs) with the central server, which are then securely aggregated over multiple data owners. Although aggregation by itself does not provably offer privacy protection, prior work showed that it may suffice if the batch size is sufficiently large. In this paper, we propose the Cocktail Party Attack (CPA) that, contrary to prior belief, is able to …
This paper investigates the problem of collecting multidimensional data throughout time (i.e., longitudinal studies) for the fundamental task of frequency estimation under Local Differential Privacy (LDP) guarantees. Contrary to frequency estimation of a single attribute, the multidimensional aspect demands particular attention to the privacy budget. Besides, when collecting user statistics longitudinally, privacy progressively degrades. Indeed, the "multiple" settings in combination (i.e., many attributes and several collections throughout time) impose several challenges, for which this paper…
