Claude Castelluccia

Raouf Kerkouche, Gergely Ács, Claude Castelluccia et al.

Federated learning becomes a prominent approach when different entities want to learn collaboratively a common model without sharing their training data. However, Federated learning has two main drawbacks. First, it is quite bandwidth inefficient as it involves a lot of message exchanges between the aggregating server and the participating entities. This bandwidth and corresponding processing costs could be prohibitive if the participating entities are, for example, mobile devices. Furthermore, although federated learning improves privacy by not sharing data, recent attacks have shown that it still leaks information about the training data. This paper presents a novel privacy-preserving federated learning scheme. The proposed scheme provides theoretical privacy guarantees, as it is based on Differential Privacy. Furthermore, it optimizes the model accuracy by constraining the model learning phase on few selected weights. Finally, as shown experimentally, it reduces the upstream and downstream bandwidth by up to 99.9% compared to standard federated learning, making it practical for mobile systems.

Raouf Kerkouche, Gergely Ács, Claude Castelluccia et al.

Federated Learning allows distributed entities to train a common model collaboratively without sharing their own data. Although it prevents data collection and aggregation by exchanging only parameter updates, it remains vulnerable to various inference and reconstruction attacks where a malicious entity can learn private information about the participants' training data from the captured gradients. Differential Privacy is used to obtain theoretically sound privacy guarantees against such inference attacks by noising the exchanged update vectors. However, the added noise is proportional to the model size which can be very large with modern neural networks. This can result in poor model quality. In this paper, compressive sensing is used to reduce the model size and hence increase model quality without sacrificing privacy. We show experimentally, using 2 datasets, that our privacy-preserving proposal can reduce the communication costs by up to 95% with only a negligible performance penalty compared to traditional non-private federated learning schemes.

Raouf Kerkouche, Gergely Ács, Claude Castelluccia

Federated Learning enables entities to collaboratively learn a shared prediction model while keeping their training data locally. It prevents data collection and aggregation and, therefore, mitigates the associated privacy risks. However, it still remains vulnerable to various security attacks where malicious participants aim at degrading the generated model, inserting backdoors, or inferring other participants' training data. This paper presents a new federated learning scheme that provides different trade-offs between robustness, privacy, bandwidth efficiency, and model accuracy. Our scheme uses biased quantization of model updates and hence is bandwidth efficient. It is also robust against state-of-the-art backdoor as well as model degradation attacks even when a large proportion of the participant nodes are malicious. We propose a practical differentially private extension of this scheme which protects the whole dataset of participating entities. We show that this extension performs as efficiently as the non-private but robust scheme, even with stringent privacy requirements but are less robust against model degradation and backdoor attacks. This suggests a possible fundamental trade-off between Differential Privacy and robustness.

Claude Castelluccia, Nataliia Bielova, Antoine Boutet et al.

This document presents an evolution of the ROBERT protocol that decentralizes most of its operations on the mobile devices. DESIRE is based on the same architecture than ROBERT but implements major privacy improvements. In particular, it introduces the concept of Private Encounter Tokens, that are secret and cryptographically generated, to encode encounters. In the DESIRE protocol, the temporary Identifiers that are broadcast on the Bluetooth interfaces are generated by the mobile devices providing more control to the users about which ones to disclose. The role of the server is merely to match PETs generated by diagnosed users with the PETs provided by requesting users. It stores minimal pseudonymous data. Finally, all data that are stored on the server are encrypted using keys that are stored on the mobile devices, protecting against data breach on the server. All these modifications improve the privacy of the scheme against malicious users and authority. However, as in the first version of ROBERT, risk scores and notifications are still managed and controlled by the server of the health authority, which provides high robustness, flexibility, and efficacy.

Benjamin Nguyen, Claude Castelluccia

In this document, we present a state of the art of anonymization techniques for classical tabular datasets. This article is geared towards a general public having some knowledge of mathematics and computer science, but with no need for specific knowledge in anonymization. The objective of this document it to explain anonymization concepts in order to be able to sanitize a dataset and compute reindentification risk. The document contains a large number of examples to help understand the calculations. ----- Dans ce document, nous présentons l'état de l'art des techniques d'anonymisation pour des bases de données classiques (i.e. des tables), à destination d'un public technique ayant une formation universitaire de base en mathématiques et informatique, mais non spécialiste. L'objectif de ce document est d'expliquer les concepts permettant de réaliser une anonymisation de données tabulaires, et de calculer les risques de réidentification. Le document est largement composé d'exemples permettant au lecteur de comprendre comment mettre en oeuvre les calculs.

Gabor Gyorgy Gulyas, Doliere Francis Some, Nataliia Bielova et al.

Recent works showed that websites can detect browser extensions that users install and websites they are logged into. This poses significant privacy risks, since extensions and Web logins that reflect user's behavior, can be used to uniquely identify users on the Web. This paper reports on the first large-scale behavioral uniqueness study based on 16,393 users who visited our website. We test and detect the presence of 16,743 Chrome extensions, covering 28% of all free Chrome extensions. We also detect whether the user is connected to 60 different websites. We analyze how unique users are based on their behavior, and find out that 54.86% of users that have installed at least one detectable extension are unique; 19.53% of users are unique among those who have logged into one or more detectable websites; and 89.23% are unique among users with at least one extension and one login. We use an advanced fingerprinting algorithm and show that it is possible to identify a user in less than 625 milliseconds by selecting the most unique combinations of extensions. Because privacy extensions contribute to the uniqueness of users, we study the trade-off between the amount of trackers blocked by such extensions and how unique the users of these extensions are. We have found that privacy extensions should be considered more useful than harmful. The paper concludes with possible countermeasures.

Gergely Acs, Luca Melis, Claude Castelluccia et al.

Generative models are used in a wide range of applications building on large amounts of contextually rich information. Due to possible privacy violations of the individuals whose data is used to train these models, however, publishing or sharing generative models is not always viable. In this paper, we present a novel technique for privately releasing generative models and entire high-dimensional datasets produced by these models. We model the generator distribution of the training data with a mixture of $k$ generative neural networks. These are trained together and collectively learn the generator distribution of a dataset. Data is divided into $k$ clusters, using a novel differentially private kernel $k$-means, then each cluster is given to separate generative neural networks, such as Restricted Boltzmann Machines or Variational Autoencoders, which are trained only on their own cluster using differentially private gradient descent. We evaluate our approach using the MNIST dataset, as well as call detail records and transit datasets, showing that it produces realistic synthetic samples, which can also be used to accurately compute arbitrary number of counting queries.

Gabor Gyorgy Gulyas, Gergely Acs, Claude Castelluccia

Several recent studies have demonstrated that people show large behavioural uniqueness. This has serious privacy implications as most individuals become increasingly re-identifiable in large datasets or can be tracked while they are browsing the web using only a couple of their attributes, called as their fingerprints. Often, the success of these attacks depend on explicit constraints on the number of attributes learnable about individuals, i.e., the size of their fingerprints. These constraints can be budget as well as technical constraints imposed by the data holder. For instance, Apple restricts the number of applications that can be called by another application on iOS in order to mitigate the potential privacy threats of leaking the list of installed applications on a device. In this work, we address the problem of identifying the attributes (e.g., smartphone applications) that can serve as a fingerprint of users given constraints on the size of the fingerprint. We give the best fingerprinting algorithms in general, and evaluate their effectiveness on several real-world datasets. Our results show that current privacy guards limiting the number of attributes that can be queried about individuals is insufficient to mitigate their potential privacy risks in many practical cases.

Jagdish Prasad Achara, Vincent Roca, Claude Castelluccia et al.

Smartphones, the devices we carry everywhere with us, are being heavily tracked and have undoubtedly become a major threat to our privacy. As "tracking the trackers" has become a necessity, various static and dynamic analysis tools have been developed in the past. However, today, we still lack suitable tools to detect, measure and compare the ongoing tracking across mobile OSs. To this end, we propose MobileAppScrutinator, based on a simple yet efficient dynamic analysis approach, that works on both Android and iOS (the two most popular OSs today). To demonstrate the current trend in tracking, we select 140 most representative Apps available on both Android and iOS AppStores and test them with MobileAppScrutinator. In fact, choosing the same set of apps on both Android and iOS also enables us to compare the ongoing tracking on these two OSs. Finally, we also discuss the effectiveness of privacy safeguards available on Android and iOS. We show that neither Android nor iOS privacy safeguards in their present state are completely satisfying.

Jagdish Prasad Achara, Javier Parra-Arnau, Claude Castelluccia

Free content and services on the Web are often supported by ads. However, with the proliferation of intrusive and privacy-invasive ads, a significant proportion of users have started to use ad blockers. As existing ad blockers are radical (they block all ads) and are not designed taking into account their economic impact, ad-based economic model of the Web is in danger today. In this paper, we target privacy-sensitive users and provide them with fine-grained control over tracking. Our working assumption is that some categories of web pages (for example, related to health, religion, etc.) are more privacy-sensitive to users than others (education, science, etc.). Therefore, our proposed approach consists in providing users with an option to specify the categories of web pages that are privacy-sensitive to them and block trackers present on such web pages only. As tracking is prevented by blocking network connections of third-party domains, we avoid not only tracking but also third-party ads. Since users will continue receiving ads on web pages belonging to non-sensitive categories, our approach essentially provides a trade-off between privacy and economy. To test the viability of our solution, we implemented it as a Google Chrome extension, named MyTrackingChoices (available on Chrome Web Store). Our real-world experiments with MyTrackingChoices show that the economic impact of ad blocking exerted by privacy-sensitive users can be significantly reduced.

Javier Parra-Arnau, Jagdish Prasad Achara, Claude Castelluccia

The intrusiveness and the increasing invasiveness of online advertising have, in the last few years, raised serious concerns regarding user privacy and Web usability. As a reaction to these concerns, we have witnessed the emergence of a myriad of ad-blocking and anti-tracking tools, whose aim is to return control to users over advertising. The problem with these technologies, however, is that they are extremely limited and radical in their approach: users can only choose either to block or allow all ads. With around 200 million people regularly using these tools, the economic model of the Web ---in which users get content free in return for allowing advertisers to show them ads--- is at serious peril. In this paper, we propose a smart Web technology that aims at bringing transparency to online advertising, so that users can make an informed and equitable decision regarding ad blocking. The proposed technology is implemented as a Web-browser extension and enables users to exert fine-grained control over advertising, thus providing them with certain guarantees in terms of privacy and browsing experience, while preserving the Internet economic model. Experimental results in a real environment demonstrate the suitability and feasibility of our approach, and provide preliminary findings on behavioral targeting from real user browsing profiles.

Jagdish Prasad Achara, Gergely Acs, Claude Castelluccia

Prior works have shown that the list of apps installed by a user reveal a lot about user interests and behavior. These works rely on the semantics of the installed apps and show that various user traits could be learnt automatically using off-the-shelf machine-learning techniques. In this work, we focus on the re-identifiability issue and thoroughly study the unicity of smartphone apps on a dataset containing 54,893 Android users collected over a period of 7 months. Our study finds that any 4 apps installed by a user are enough (more than 95% times) for the re-identification of the user in our dataset. As the complete list of installed apps is unique for 99% of the users in our dataset, it can be easily used to track/profile the users by a service such as Twitter that has access to the whole list of installed apps of users. As our analyzed dataset is small as compared to the total population of Android users, we also study how unicity would vary with larger datasets. This work emphasizes the need of better privacy guards against collection, use and release of the list of installed apps.

Minh-Dung Tran, Gergely Acs, Claude Castelluccia

Retargeting ads are increasingly prevalent on the Internet as their effectiveness has been shown to outperform conventional targeted ads. Retargeting ads are not only based on users' interests, but also on their intents, i.e. commercial products users have shown interest in. Existing retargeting systems heavily rely on tracking, as retargeting companies need to know not only the websites a user has visited but also the exact products on these sites. They are therefore very intrusive, and privacy threatening. Furthermore, these schemes are still sub-optimal since tracking is partial, and they often deliver ads that are obsolete (because, for example, the targeted user has already bought the advertised product). This paper presents the first privacy-preserving retargeting ads system. In the proposed scheme, the retargeting algorithm is distributed between the user and the advertiser such that no systematic tracking is necessary, more control and transparency is provided to users, but still a lot of targeting flexibility is provided to advertisers. We show that our scheme, that relies on homomorphic encryption, can be efficiently implemented and trivially solves many problems of existing schemes, such as frequency capping and ads freshness.

Claude Castelluccia, Abdelberi Chaabane, Markus Dürmuth et al.

Passwords are widely used for user authentication and, despite their weaknesses, will likely remain in use in the foreseeable future. Human-generated passwords typically have a rich structure, which makes them susceptible to guessing attacks. In this paper, we study the effectiveness of guessing attacks based on Markov models. Our contributions are two-fold. First, we propose a novel password cracker based on Markov models, which builds upon and extends ideas used by Narayanan and Shmatikov (CCS 2005). In extensive experiments we show that it can crack up to 69% of passwords at 10 billion guesses, more than all probabilistic password crackers we compared again t. Second, we systematically analyze the idea that additional personal information about a user helps in speeding up password guessing. We find that, on average and by carefully choosing parameters, we can guess up to 5% more passwords, especially when the number of attempts is low. Furthermore, we show that the gain can go up to 30% for passwords that are actually based on personal attributes. These passwords are clearly weaker and should be avoided. Our cracker could be used by an organization to detect and reject them. To the best of our knowledge, we are the first to systematically study the relationship between chosen passwords and users' personal information. We test and validate our results over a wide collection of leaked password databases.

Gergely Acs, Claude Castelluccia

This paper presents a new privacy-preserving smart metering system. Our scheme is private under the differential privacy model and therefore provides strong and provable guarantees. With our scheme, an (electricity) supplier can periodically collect data from smart meters and derive aggregated statistics while learning only limited information about the activities of individual households. For example, a supplier cannot tell from a user's trace when he watched TV or turned on heating. Our scheme is simple, efficient and practical. Processing cost is very limited: smart meters only have to add noise to their data and encrypt the results with an efficient stream cipher.