Federated learning (FL) is a machine learning setting where many clients (e.g. mobile devices or whole organizations) collaboratively train a model under the orchestration of a central server (e.g. service provider), while keeping the training data decentralized. FL embodies the principles of focused data collection and minimization, and can mitigate many of the systemic privacy risks and costs resulting from traditional, centralized machine learning and data science approaches. Motivated by the explosive growth in FL research, this paper discusses recent advances and presents an extensive collection of open problems and challenges.

Advances and Open Problems in Federated Learning

Sensitive statistics are often collected across sets of users, with repeated collection of reports done over time. For example, trends in users' private preferences or software usage may be monitored via such reports. We study the collection of such statistics in the local differential privacy (LDP) model, and describe an algorithm whose privacy cost is polylogarithmic in the number of changes to a user's value. 
More fundamentally---by building on anonymity of the users' reports---we also demonstrate how the privacy cost of our LDP algorithm can actually be much lower when viewed in the central model of differential privacy. We show, via a new and general privacy amplification technique, that any permutation-invariant algorithm satisfying $\varepsilon$-local differential privacy will satisfy $(O(\varepsilon \sqrt{\log(1/\delta)/n}), \delta)$-central differential privacy. By this, we explain how the high noise and $\sqrt{n}$ overhead of LDP protocols is a consequence of them being significantly more private in the central model. As a practical corollary, our results imply that several LDP-based industrial deployments may have much lower privacy cost than their advertised $\varepsilon$ would indicate---at least if reports are anonymized.

Amplification by Shuffling: From Local to Central Differential Privacy via Anonymity

Local differential privacy (LDP) is a recently proposed privacy standard for collecting and analyzing data, which has been used, e.g., in the Chrome browser, iOS and macOS. In LDP, each user perturbs her information locally, and only sends the randomized version to an aggregator who performs analyses, which protects both the users and the aggregator against private information leaks. Although LDP has attracted much research attention in recent years, the majority of existing work focuses on applying LDP to complex data and/or analysis tasks. In this paper, we point out that the fundamental problem of collecting multidimensional data under LDP has not been addressed sufficiently, and there remains much room for improvement even for basic tasks such as computing the mean value over a single numeric attribute under LDP. Motivated by this, we first propose novel LDP mechanisms for collecting a numeric attribute, whose accuracy is at least no worse (and usually better) than existing solutions in terms of worst-case noise variance. Then, we extend these mechanisms to multidimensional data that can contain both numeric and categorical attributes, where our mechanisms always outperform existing solutions regarding worst-case noise variance. As a case study, we apply our solutions to build an LDP-compliant stochastic gradient descent algorithm (SGD), which powers many important machine learning tasks. Experiments using real datasets confirm the effectiveness of our methods, and their advantages over existing solutions.

Collecting and Analyzing Multidimensional Data with Local Differential Privacy

Secure aggregation is a cryptographic primitive that enables a server to learn the sum of the vector inputs of many clients. Bonawitz et al. (CCS 2017) presented a construction that incurs computation and communication for each client linear in the number of parties. While this functionality enables a broad range of privacy preserving computational tasks, scaling concerns limit its scope of use. We present the first constructions for secure aggregation that achieve polylogarithmic communication and computation per client. Our constructions provide security in the semi-honest and the semi-malicious settings where the adversary controls the server and a δ-fraction of the clients, and correctness with up to δ-fraction dropouts among the clients. Our constructions show how to replace the complete communication graph of Bonawitz et al., which entails the linear overheads, with a k-regular graph of logarithmic degree while maintaining the security guarantees. Beyond improving the known asymptotics for secure aggregation, our constructions also achieve very efficient concrete parameters. The semi-honest secure aggregation can handle a billion clients at the per-client cost of the protocol of Bonawitz et al. for a thousand clients. In the semi-malicious setting with 10 4 clients, each client needs to communicate only with 3% of the clients to have a guarantee that its input has been added together with the inputs of at least 5000 other clients, while withstanding up to 5% corrupt clients and 5% dropouts. We also show an application of secure aggregation to the task of secure shuffling which enables the first cryptographically secure instantiation of the shuffle model of differential privacy.

https://dl.acm.org/doi/pdf/10.1145/3372297.3417885

Secure Single-Server Aggregation with (Poly)Logarithmic Overhead

This work studies differential privacy in the context of the recently proposed shuffle model. Unlike in the local model, where the server collecting privatized data from users can track back an input to a specific user, in the shuffle model users submit their privatized inputs to a server anonymously. This setup yields a trust model which sits in between the classical curator and local models for differential privacy. The shuffle model is the core idea in the Encode, Shuffle, Analyze (ESA) model introduced by Bittau et al. (SOPS 2017). Recent work by Cheu et al. (EUROCRYPT 2019) analyzes the differential privacy properties of the shuffle model and shows that in some cases shuffled protocols provide strictly better accuracy than local protocols. Additionally, Erlingsson et al. (SODA 2019) provide a privacy amplification bound quantifying the level of curator differential privacy achieved by the shuffle model in terms of the local differential privacy of the randomizer used by each user.

The Privacy Blanket of the Shuffle Model

The large-scale monitoring of computer users' software activities has become commonplace, e.g., for application telemetry, error reporting, or demographic profiling. This paper describes a principled systems architecture---Encode, Shuffle, Analyze (ESA)---for performing such monitoring with high utility while also protecting user privacy. The ESA design, and its Prochlo implementation, are informed by our practical experiences with an existing, large deployment of privacy-preserving software monitoring. With ESA, the privacy of monitored users' data is guaranteed by its processing in a three-step pipeline. First, the data is encoded to control scope, granularity, and randomness. Second, the encoded data is collected in batches subject to a randomized threshold, and blindly shuffled, to break linkability and to ensure that individual data items get "lost in the crowd" of the batch. Third, the anonymous, shuffled data is analyzed by a specific analysis engine that further prevents statistical inference attacks on analysis results. ESA extends existing best-practice methods for sensitive-data analytics, by using cryptography and statistical techniques to make explicit how data is elided and reduced in precision, how only common-enough, anonymous data is analyzed, and how this is done for only specific, permitted purposes. As a result, ESA remains compatible with the established workflows of traditional database analysis. Strong privacy guarantees, including differential privacy, can be established at each processing step to defend against malice or compromise at one or more of those steps. Prochlo develops new techniques to harden those steps, including the Stash Shuffle, a novel scalable and efficient oblivious-shuffling algorithm based on Intel's SGX, and new applications of cryptographic secret sharing and blinding. We describe ESA and Prochlo, as well as experiments that validate their ability to balance utility and privacy.

https://dl.acm.org/doi/pdf/10.1145/3132747.3132769

Prochlo: Strong Privacy for Analytics in the Crowd

The large-scale monitoring of computer users' software activities has become commonplace, e.g., for application telemetry, error reporting, or demographic profiling. This paper describes a principled systems architecture---Encode, Shuffle, Analyze (ESA)---for performing such monitoring with high utility while also protecting user privacy. The ESA design, and its Prochlo implementation, are informed by our practical experiences with an existing, large deployment of privacy-preserving software monitoring. 
(cont.; see the paper)

Some embodiments provide a system that restricts privileges for a target process in an operating system. During operation, the system receives a request from the target process to restrict the privileges. Next, the system executes a binary that processes the request using services provided by a kernel of the operating system.

Julien Tinnes

Papers

Prochlo: Strong Privacy for Analytics in the Crowd

Prochlo: Strong Privacy for Analytics in the Crowd

Restricting privileges of first privileged process in operating system using second privileged process