As data are increasingly being stored in different silos and societies becoming more aware of data privacy issues, the traditional centralized training of artificial intelligence (AI) models is facing efficiency and privacy challenges. Recently, federated learning (FL) has emerged as an alternative solution and continue to thrive in this new reality. Existing FL protocol design has been shown to be vulnerable to adversaries within or outside of the system, compromising data privacy and system robustness. Besides training powerful global models, it is of paramount importance to design FL systems that have privacy guarantees and are resistant to different types of adversaries. In this paper, we conduct the first comprehensive survey on this topic. Through a concise introduction to the concept of FL, and a unique taxonomy covering: 1) threat models; 2) poisoning attacks and defenses against robustness; 3) inference attacks and defenses against privacy, we provide an accessible review of this important topic. We highlight the intuitions, key techniques as well as fundamental assumptions adopted by various attacks and defenses. Finally, we discuss promising future research directions towards robust and privacy-preserving federated learning.

Privacy and Robustness in Federated Learning: Attacks and Defenses.

Federated learning (FL) is an emerging distributed machine learning paradigm which addresses critical data privacy issues in machine learning by enabling clients, using an aggregation server (aggregator), to jointly train a global model without revealing their training data. Thereby, it improves not only privacy but is also efficient as it uses the computation power and data of potentially millions of clients for training in parallel. However, FL is vulnerable to so-called inference attacks by malicious aggregators which can infer information about clients’ data from their model updates. Secure aggregation restricts the central aggregator to only learn the summation or average of the updates of clients. Unfortunately, existing protocols for secure aggregation for FL suffer from high communication, computation, and many communication rounds.In this work, we present SAFELearn, a generic design for efficient private FL systems that protects against inference attacks that have to analyze individual clients’ model updates using secure aggregation. It is flexibly adaptable to the efficiency and security requirements of various FL applications and can be instantiated with MPC or FHE. In contrast to previous works, we only need 2 rounds of communication in each training iteration, do not use any expensive cryptographic primitives on clients, tolerate dropouts, and do not rely on a trusted third party. We implement and benchmark an instantiation of our generic design with secure two-party computation. Our implementation aggregates 500 models with more than 300K parameters in less than 0.5 seconds.

SAFELearn: Secure Aggregation for private FEderated Learning

Decentralized paradigm in the field of cybersecurity and machine learning (ML) for the emerging Internet of Things (IoT) has gained a lot of attention from the government, academia, and industries in recent years. Federated cybersecurity (FC) is regarded as a revolutionary concept to make the IoT safer and more efficient in the future. This emerging concept has the potential of detecting security threats, taking countermeasures, and limiting the spreading of threats over the IoT network system efficiently. An objective of cybersecurity is achieved by forming the federation of the learned and shared model on top of various participants. Federated learning (FL), which is regarded as a privacy-aware ML model, is particularly useful to secure the vulnerable IoT environment. In this article, we start with background and comparison of centralized learning, distributed on-site learning, and FL, which is then followed by a survey of the application of FL to cybersecurity for IoT. This survey primarily focuses on the security aspect but it also discusses several approaches that address the performance issues (e.g., accuracy, latency, resource constraint, and others) associated with FL, which may impact the security and overall performance of the IoT. To anticipate the future evolution of this new paradigm, we discuss the main ongoing research efforts, challenges, and research trends in this area. With this article, readers can have a more thorough understanding of FL for cybersecurity as well as cybersecurity for FL, different security attacks, and countermeasures.

Recent Advances on Federated Learning for Cybersecurity and Cybersecurity for Federated Learning for Internet of Things

Recent advances in communication technologies and the Internet-of-Medical-Things (IOMT) have transformed smart healthcare enabled by artificial intelligence (AI). Traditionally, AI techniques require centralized data collection and processing that may be infeasible in realistic healthcare scenarios due to the high scalability of modern healthcare networks and growing data privacy concerns. Federated Learning (FL), as an emerging distributed collaborative AI paradigm, is particularly attractive for smart healthcare, by coordinating multiple clients (e.g., hospitals) to perform AI training without sharing raw data. Accordingly, we provide a comprehensive survey on the use of FL in smart healthcare. First, we present the recent advances in FL, the motivations, and the requirements of using FL in smart healthcare. The recent FL designs for smart healthcare are then discussed, ranging from resource-aware FL, secure and privacy-aware FL to incentive FL and personalized FL. Subsequently, we provide a state-of-the-art review on the emerging applications of FL in key healthcare domains, including health data management, remote health monitoring, medical imaging, and COVID-19 detection. Several recent FL-based smart healthcare projects are analyzed, and the key lessons learned from the survey are also highlighted. Finally, we discuss interesting research challenges and possible directions for future FL research in smart healthcare. 

/pdf/federated-learning-for-smart-healthcare-a-survey-2kuzq1dr.pdf

Federated Learning for Smart Healthcare: A Survey

Recent studies have shown that federated learning (FL) is vulnerable to poisoning attacks that inject a backdoor into the global model. These attacks are effective even when performed by a single client, and undetectable by most existing defensive techniques. In this paper, we propose Backdoor detection via Feedback-based Federated Learning (BAFFLE), a novel defense to secure FL against backdoor attacks. The core idea behind BAFFLE is to leverage data of multiple clients not only for training but also for uncovering model poisoning. We exploit the availability of diverse datasets at the various clients by incorporating a feedback loop into the FL process, to integrate the views of those clients when deciding whether a given model update is genuine or not. We show that this powerful construct can achieve very high detection rates against state-of-the-art backdoor attacks, even when relying on straightforward methods to validate the model. Through empirical evaluation using the CIFAR-10 and FEMNIST datasets, we show that by combining the feedback loop with a method that suspects poisoning attempts by assessing the per-class classification performance of the updated model, BAFFLE reliably detects state-of-the-art backdoor attacks with a detection accuracy of 100% and a false-positive rate below 5%. Moreover, we show that our solution can detect adaptive attacks aimed at bypassing the defense.

BaFFLe: Backdoor detection via Feedback-based Federated Learning.

BaFFLe: Backdoor Detection via Feedback-based Federated Learning

Balancing Quality and Efficiency in Private Clustering with Affinity Propagation

Modeling the spread of COVID-19 is crucial for any effort to manage the pandemic. However, detailed epidemiological simulations suffer from a scarcity of relevant empirical data, such as social contact graphs, because such data is inherently privacy-critical. Thus, there is an urgent need for a method to perform powerful epidemiological simulations on real-world contact graphs without disclosing privacy-critical information. In this work, we propose a practical framework for privacypreserving epidemiological modeling (PEM) on contact information stored on mobile phones, like the ones collected by already deployed contact tracing apps. Unlike those apps, PEM allows for meaningful epidemiological simulations. This is enabled by a novel Threshold-PIR-SUM protocol to privately retrieve the sum of a fixed number of distinct values without revealing individual values. PEM protects the privacy of the users by not revealing sensitive data to the system operator or other participants, while enabling detailed predictive models of pandemic spread.

/pdf/pem-privacy-preserving-epidemiological-modeling-4zwc41m2wr.pdf

Helen Möllering

Papers

SAFELearn: Secure Aggregation for private FEderated Learning

BaFFLe: Backdoor detection via Feedback-based Federated Learning.

BaFFLe: Backdoor Detection via Feedback-based Federated Learning

Balancing Quality and Efficiency in Private Clustering with Affinity Propagation

PEM: Privacy-preserving Epidemiological Modeling.