Wireless Sensor Networks (WSNs) rely on in-network aggregation for efficiency, however, this comes at a price: A single adversary can severely influence the outcome by contributing an arbitrary partial aggregate value. Secure in-network aggregation can detect such manipulation. But as long as such faults persist, no aggregation result can be obtained. In contrast, the collection of individual sensor node values is robust and solves the problem of availability, yet in an inefficient way. Our work seeks to bridge this gap in secure data collection: We propose a system that enhances availability with an efficiency close to that of in-network aggregation. To achieve this, our scheme relies on costly operations to localize and exclude nodes that manipulate the aggregation, but \emph{only} when a failure is detected. The detection of aggregation disruptions and the removal of faulty nodes provides robustness. At the same time, after removing faulty nodes, the WSN can enjoy low cost (secure) aggregation. Thus, the high exclusion cost is amortized, and efficiency increases.
Secure aggregation is a cryptographic protocol that securely computes the aggregation of its inputs. It is pivotal in keeping model updates private in federated learning. Indeed, the use of secure aggregation prevents the server from learning the value and the source of the individual model updates provided by the users, hampering inference and data attribution attacks. In this work, we show that a malicious server can easily elude secure aggregation as if the latter were not in place. We devise two different attacks capable of inferring information on individual private training datasets, ind…
We propose and experimentally evaluate a novel secure aggregation algorithm targeted at cross-organizational federated learning applications with a fixed set of participating learners. Our solution organizes learners in a chain and encrypts all traffic to reduce the controller of the aggregation to a mere message broker. We show that our algorithm scales better and is less resource demanding than existing solutions, while being easy to implement on constrained platforms. With 36 nodes our method outperforms state-of-the-art secure aggregation by 70x, and 56x with and without failover, respecti…
Secure Multiparty Computation (MPC) can improve the security and privacy of data owners while allowing analysts to perform high quality analytics. Secure aggregation is a secure distributed mechanism to support federated deep learning without the need for trusted third parties. In this paper we present a highly performant secure aggregation protocol with sub-linear communication complexity. Our protocol achieves greater communication and computation efficiencies through a group-based approach. It is similar to secret sharing protocols extended to vectors of values-aka gradients-but within grou…
We propose a secure data aggregation (SDA) framework for mobile sensor networks whose topology changes dynamically with time. The SDA framework (designed to be resilient to both insider and outsider attacks) comprises of a pair-wise key establishment mechanism run along the edges of a data gathering tree and a distributed trust evaluation model that is tightly integrated with the data aggregation process itself. If an aggregator node already shares a secret key with its child node, the two nodes locally coordinate to refresh and establish a new pair-wise secret key; otherwise, the aggregator n…
Secure aggregation is a popular protocol in privacy-preserving federated learning, which allows model aggregation without revealing the individual models in the clear. On the other hand, conventional secure aggregation protocols incur a significant communication overhead, which can become a major bottleneck in real-world bandwidth-limited applications. Towards addressing this challenge, in this work we propose a lightweight gradient sparsification framework for secure aggregation, in which the server learns the aggregate of the sparsified local model updates from a large number of users, but w…
Secure Aggregation protocols allow a collection of mutually distrust parties, each holding a private value, to collaboratively compute the sum of those values without revealing the values themselves. We consider training a deep neural network in the Federated Learning model, using distributed stochastic gradient descent across user-held training data on mobile devices, wherein Secure Aggregation protects each user's model gradient. We design a novel, communication-efficient Secure Aggregation protocol for high-dimensional data that tolerates up to 1/3 users failing to complete the protocol. Fo…
Federated learning is a collaborative method that aims to preserve data privacy while creating AI models. Current approaches to federated learning tend to rely heavily on secure aggregation protocols to preserve data privacy. However, to some degree, such protocols assume that the entity orchestrating the federated learning process (i.e., the server) is not fully malicious or dishonest. We investigate vulnerabilities to secure aggregation that could arise if the server is fully malicious and attempts to obtain access to private, potentially sensitive data. Furthermore, we provide a method to f…
Secure aggregation is widely used in horizontal Federated Learning (FL), to prevent leakage of training data when model updates from data owners are aggregated. Secure aggregation protocols based on Homomorphic Encryption (HE) have been utilized in industrial cross-silo FL systems, one of the settings involved with privacy-sensitive organizations such as financial or medical, presenting more stringent requirements on privacy security. However, existing HE-based solutions have limitations in efficiency and security guarantees against colluding adversaries without a Trust Third Party. This paper…
Secure aggregation, which is a core component of federated learning, aggregates locally trained models from distributed users at a central server. The ``secure'' nature of such aggregation consists of the fact that no information about the local users' data must be leaked to the server except the aggregated local models. In order to guarantee security, some keys may be shared among the users. After the key sharing phase, each user masks its trained model which is then sent to the server. This paper follows the information theoretic secure aggregation problem originally formulated by Zhao and S…
Secure linear aggregation is to linearly aggregate private inputs of different users with privacy protection. The server in a federated learning (FL) environment can fulfill any linear computation on private inputs of users through the secure linear aggregation. At present, based on pseudo-random number generator and one-time padding technique, one can efficiently compute the sum of user inputs in FL, but linear calculations of user inputs are not well supported. Based on decentralized threshold additive homomorphic encryption (DTAHE) schemes, this paper provides a secure linear aggregation pr…
Federated Learning enables one to jointly train a machine learning model across distributed clients holding sensitive datasets. In real-world settings, this approach is hindered by expensive communication and privacy concerns. Both of these challenges have already been addressed individually, resulting in competing optimisations. In this article, we tackle them simultaneously for one of the first times. More precisely, we adapt compression-based federated techniques to additive secret sharing, leading to an efficient secure aggregation protocol, with an adaptable security level. We prove its p…
Google（谷歌）；California Institute of Technology（加州理工学院）
Federated Learning enables mobile devices to collaboratively learn a shared inference model while keeping all the training data on a user's device, decoupling the ability to do machine learning from the need to store the data in the cloud. Existing work on federated learning with limited communication demonstrates how random rotation can enable users' model updates to be quantized much more efficiently, reducing the communication cost between users and the server. Meanwhile, secure aggregation enables the server to learn an aggregate of at least a threshold number of device's model contributio…
Privacy preserving multi-party computation has many applications in areas such as medicine and online advertisements. In this work, we propose a framework for distributed, secure machine learning among untrusted individuals. The framework consists of two parts: a two-step training protocol based on homomorphic addition and a zero knowledge proof for data validity. By combining these two techniques, our framework provides privacy of per-user data, prevents against a malicious user contributing corrupted data to the shared pool, enables each user to self-compute the results of the algorithm with…
University of Vermont（佛蒙特大学）；MassMutual（万通互惠）；Raytheon（雷神公司）
Federated machine learning leverages edge computing to develop models from network user data, but privacy in federated learning remains a major challenge. Techniques using differential privacy have been proposed to address this, but bring their own challenges -- many require a trusted third party or else add too much noise to produce useful models. Recent advances in \emph{secure aggregation} using multiparty computation eliminate the need for a third party, but are computationally expensive especially at scale. We present a new federated learning protocol that leverages a novel differentially…
Federated learning enables multiple data owners to jointly train a machine learning model without revealing their private datasets. However, a malicious aggregation server might use the model parameters to derive sensitive information about the training dataset used. To address such leakage, differential privacy and cryptographic techniques have been investigated in prior work, but these often result in large communication overheads or impact model performance. To mitigate this centralization of power, we propose SCOTCH, a decentralized m-party secure-computation framework for federated aggreg…
Although homomorphic encryption can be incorporated into neural network layers for securing machine learning tasks, such as confidential inference over encrypted data samples and encrypted local models in federated learning, the computational overhead has been an Achilles heel. This paper proposes a caching protocol, namely CHEM, such that tensor ciphertexts can be constructed from a pool of cached radixes rather than carrying out expensive encryption operations. From a theoretical perspective, we demonstrate that CHEM is semantically secure and can be parameterized with straightforward analys…
University of Southern California（南加州大学）；University of California, Riverside（加州大学河滨分校）
Federated learning is a distributed framework for training machine learning models over the data residing at mobile devices, while protecting the privacy of individual users. A major bottleneck in scaling federated learning to a large number of users is the overhead of secure model aggregation across many users. In particular, the overhead of the state-of-the-art protocols for secure model aggregation grows quadratically with the number of users. In this paper, we propose the first secure aggregation framework, named Turbo-Aggregate, that in a network with $N$ users achieves a secure aggregati…
In [3], the authors proposed a highly efficient secure and privacy-preserving scheme for secure vehicular communications. The proposed scheme consists of four protocols: system setup, protocol for STP and STK distribution, protocol for common string synchronization, and protocol for vehicular communications. Here we define the security models for the protocol for STP and STK distribution, and the protocol for vehicular communications,respectively. We then prove that these two protocols are secure in our models.
The pervasive adoption of Internet-connected digital services has led to a growing concern in the personal data privacy of their customers. On the other hand, machine learning (ML) techniques have been widely adopted by digital service providers to improve operational productivity and customer satisfaction. ML inevitably accesses and processes users' personal data, which could potentially breach the relevant privacy protection regulations if not performed carefully. The situation is exacerbated by the cloud-based implementation of digital services when user data are captured and stored in dist…
We consider the generic problem of Secure Aggregation of Distributed Information (SADI), where several agents acting as a team have information distributed among them, modeled by means of a publicly known deck of cards distributed among the agents, so that each of them knows only her cards. The agents have to exchange and aggregate the information about how the cards are distributed among them by means of public announcements over insecure communication channels, intercepted by an adversary "eavesdropper", in such a way that the adversary does not learn who holds any of the cards. We present a…
Coordinated Sciences Laboratory（协同科学实验室）；University of Illinois at Urbana-Champaign（伊利诺伊大学厄巴纳-香槟分校）
Federated learning systems that jointly preserve Byzantine robustness and privacy have remained an open problem. Robust aggregation, the standard defense for Byzantine attacks, generally requires server access to individual updates or nonlinear computation -- thus is incompatible with privacy-preserving methods such as secure aggregation via multiparty computation. To this end, we propose SHARE (Secure Hierarchical Robust Aggregation), a distributed learning framework designed to cryptographically preserve client update privacy and robustness to Byzantine adversaries simultaneously. The key id…
Two recent papers describe almost exactly the same group key establishment protocol for wireless sensor networks. Quite part from the duplication issue, we show that both protocols are insecure and should not be used - a member of a group can successfully impersonate the key generation centre and persuade any other group member to accept the wrong key value. This breaks the stated objectives of the schemes.
Aggregation of values that need to be kept confidential while guaranteeing the robustness of the process and the correctness of the result is required in an increasing number of applications. We propose an aggregation algorithm, which supports a large spectrum of potential applications including complex voting protocols. It relies on the distributed hash table Kademlia, used in BitTorrent, for pseudonymous communication between randomly predetermined peers to ensure a high degree of confidentiality which does not solely relies on cryptography. The distribution of data and computation limits th…
In the robust secure aggregation problem, a server wishes to learn and only learn the sum of the inputs of a number of users while some users may drop out (i.e., may not respond). The identity of the dropped users is not known a priori and the server needs to securely recover the sum of the remaining surviving users. We consider the following minimal two-round model of secure aggregation. Over the first round, any set of no fewer than $U$ users out of $K$ users respond to the server and the server wants to learn the sum of the inputs of all responding users. The remaining users are viewed as d…
