Federated Learning (FL) enables collaborative model training while keeping client data private. However, exposing individual client updates makes FL vulnerable to reconstruction attacks. Secure aggregation mitigates such privacy risks but prevents the server from verifying the validity of each client update, creating a privacy-robustness tradeoff. Recent efforts attempt to address this tradeoff by enforcing checks on client updates using zero-knowledge proofs, but they support limited predicates and often depend on public validation data. We propose SLVR, a general framework that securely leverages clients' private data through secure multi-party computation. By utilizing clients' data, SLVR not only eliminates the need for public validation data, but also enables a wider range of checks for robustness, including cross-client accuracy validation. It also adapts naturally to distribution shifts in client data as it can securely refresh its validation data up-to-date. Our empirical evaluations show that SLVR improves robustness against model poisoning attacks, particularly outperforming existing methods by up to 50% under adaptive attacks. Additionally, SLVR demonstrates effective adaptability and stable convergence under various distribution shift scenarios.
FedPoP: Federated Learning Meets Proof of Participation
IMDEA Networks Institute（IMDEA网络研究所）；Universidad Carlos III de Madrid（卡洛斯三世大学）；University of California, Irvine（加州大学尔湾分校）；Stealth Software Technologies Inc.（隐形软件技术公司）
Cloud storage services like Dropbox and Google Drive are widely used by individuals and businesses. Two attractive features of these services are 1) the automatic synchronization of files between multiple client devices and 2) the possibility to share files with other users. However, privacy of cloud data is a growing concern for both individuals and businesses. Encrypting data on the client-side before uploading it is an effective privacy safeguard, but it requires all client devices to have the decryption key. Current solutions derive these keys solely from user-chosen passwords, which have …
Cloud computing enables clients with limited computational power to economically outsource their large scale computations to a public cloud with huge computational power. Cloud has the massive storage, computational power and software which can be used by clients for reducing their computational overhead and storage limitation. But in case of outsourcing, privacy of client's confidential data must be maintained. We have designed a protocol for outsourcing large scale Eigen value problem to a malicious cloud which provides input/output data security, result verifiability and client's efficiency…
ConcurORAM is a parallel, multi-client oblivious RAM (ORAM) that eliminates waiting for concurrent stateless clients and allows overall throughput to scale gracefully, without requiring trusted third party components (proxies) or direct inter-client coordination. A key insight behind ConcurORAM is the fact that, during multi-client data access, only a subset of the concurrently-accessed server-hosted data structures require access privacy guarantees. Everything else can be safely implemented as oblivious data structures that are later synced securely and efficiently during an ORAM "eviction". …
FedSiKD: Clients Similarity and Knowledge Distillation: Addressing Non-i.i.d. and Constraints in Federated Learning
King Abdulaziz University（阿卜杜勒阿齐兹国王大学）；Lun Startup Studio（伦创业工作室）；Dhirubhai Ambani Institute of Information and Communication Technology（迪鲁拜·安巴尼信息与通信技术学院）；Southern Illinois University at Carbondale（南伊利诺伊大学卡本代尔分校）；Riphah International University Islamabad（里法国际大学伊斯兰堡分校）
End-to-end encryption (E2EE) by messaging platforms enable people to securely and privately communicate with one another. Its widespread adoption however raised concerns that illegal content might now be shared undetected. Following the global pushback against key escrow systems, client-side scanning based on perceptual hashing has been recently proposed by tech companies, governments and researchers to detect illegal content in E2EE communications. We here propose the first framework to evaluate the robustness of perceptual hashing-based client-side scanning to detection avoidance attacks and…
University of California, Berkeley（加州大学伯克利分校）；Hong Kong University of Science and Technology（香港科技大学）
Prior efforts in enhancing federated learning (FL) security fall into two categories. At one end of the spectrum, some work uses secure aggregation techniques to hide the individual client's updates and only reveal the aggregated global update to a malicious server that strives to infer the clients' privacy from their updates. At the other end of the spectrum, some work uses Byzantine-robust FL protocols to suppress the influence of malicious clients' updates. We present a federated learning protocol F2ED-LEARNING, which, for the first time, offers bidirectional defense to simultaneously comba…
Multi-center Medical Data Mining with FL-Net - A One-stop Shop for Federated Learning
University of Hamburg（汉堡大学）；Sapienza University of Rome（罗马第一大学）；University of Campania Luigi Vanvitelli（坎帕尼亚大学路易吉·万维泰利）；Semmelweis University（塞麦尔维斯大学）；SC Egnosis SRL（SC Egnosis公司）；UCL Queen Square Institute of Neurology（伦敦大学学院皇后广场神经学研究所）；LMU München（慕尼黑大学）；LMU Hospital（慕尼黑大学医院）；Research Institute – Digital Human Rights Center（数字人权中心研究所）；Medical University of Vienna（维也纳医科大学）；SBA Research gGmbH（SBA研究中心）；Joslin Diabetes Center（乔斯林糖尿病中心）；Harvard Medical School（哈佛医学院）；Karolinska Institutet（卡罗林斯卡学院）；Södersjukhuset（南斯德哥尔摩综合医院）；University College Cork（科克大学学院）；Karolinska University Hospital（卡罗林斯卡大学医院）；Mass General Brigham（麻省总布里格姆医院）；University of Southern Denmark（南丹麦大学）；Université Paris-Saclay（巴黎-萨克雷大学）；INRAE（法国国家农业、食品与环境研究院）
This paper proposes DeepSecure, a novel framework that enables scalable execution of the state-of-the-art Deep Learning (DL) models in a privacy-preserving setting. DeepSecure targets scenarios in which neither of the involved parties including the cloud servers that hold the DL model parameters or the delegating clients who own the data is willing to reveal their information. Our framework is the first to empower accurate and scalable DL analysis of data generated by distributed clients without sacrificing the security to maintain efficiency. The secure DL computation in DeepSecure is perform…
