机器学习（ML）的进步显著革新了医学图像分析，促使医院依赖外部 ML 服务。然而，与第三方共享胸部 X 光等敏感患者数据会带来内在隐私风险。为解决这一问题，我们提出 MedBlindTuner，这是一个利用全同态加密（FHE）和数据高效图像 Transformer（DEiT）的隐私保护框架。MedBlindTuner 使得 ML 模型能够完全在 FHE 加密的医学图像上进行训练。我们的实验评估表明，MedBlindTuner 达到了与在非加密图像上训练的模型相当的准确率，为外包 ML 计算提供了一种安全解决方案，同时保护患者数据隐私。据我们所知，这是该领域首个使用数据高效图像 Transformer 和全同态加密的工作。
英文摘要：
Advancements in machine learning (ML) have significantly revolutionized medical image analysis, prompting hospitals to rely on external ML services. However, the exchange of sensitive patient data, such as chest X-rays, poses inherent privacy risks when shared with third parties. Addressing this concern, we propose MedBlindTuner, a privacy-preserving framework leveraging fully homomorphic encryption (FHE) and a data-efficient image transformer (DEiT). MedBlindTuner enables the training of ML models exclusively on FHE-encrypted medical images. Our experimental evaluation demonstrates that MedBlindTuner achieves comparable accuracy to models trained on non-encrypted images, offering a secure solution for outsourcing ML computations while preserving patient data privacy. To the best of our knowledge, this is the first work that uses data-efficient image transformers and fully homomorphic encryption in this domain.
In this technical report, we explore the use of homomorphic encryption (HE) in the context of training and predicting with deep learning (DL) models to deliver strict \textit{Privacy by Design} services, and to enforce a zero-trust model of data governance. First, we show how HE can be used to make predictions over medical images while preventing unauthorized secondary use of data, and detail our results on a disease classification task with OCT images. Then, we demonstrate that HE can be used to secure the training of DL models through federated learning, and report some experiments using 3D …
European Research Council Executive Agency（欧洲研究委员会执行局）；Vrije Universiteit Brussel（布鲁塞尔自由大学）
Homomorphic encryption aims at allowing computations on encrypted data without decryption other than that of the final result. This could provide an elegant solution to the issue of privacy preservation in data-based applications, such as those using machine learning, but several open issues hamper this plan. In this work we assess the possibility for homomorphic encryption to fully implement its program without relying on other techniques, such as multiparty computation (SMPC), which may be impossible in many use cases (for instance due to the high level of communication required). We proceed…
Fully homomorphic encryption is a promising crypto primitive to encrypt your data while allowing others to compute on the encrypted data. But there are many well-known problems with fully homomorphic encryption such as CCA security and circuit privacy problem. Despite these problems, there are still many companies are currently using or preparing to use fully homomorphic encryption to build data security applications. It seems that the full homomorphic encryption is very close to practicality and these problems can be easily mitigated in implementation. Although the those problems are well kno…
This article presents the application of homomorphic authenticators, replication encodings to be precise, to multigroup fully homomorphic encryption schemes. Following the works of Gennaro and Wichs on homomorphic authenticators in combination with the work of multigroup schemes by Kwak et al. we present a verifiable solution for a fully homomorphic primitive that includes the multikey, multiparty and single-user cases. Furthermore, we propose a line of prospective research in constrained-resource scenarios.
We propose a multi-bit leveled fully homomorphic encryption scheme using multivariate polynomial evaluations. The security of the scheme depends on the hardness of the Learning with Errors (LWE) problem. For homomorphic multiplication, the scheme uses a polynomial based technique that does not require relinearization (and key switching). The noise associated with the ciphertext increases only linearly with every multiplication.
In this paper we provide a survey of various libraries for homomorphic encryption. We describe key features and trade-offs that should be considered while choosing the right approach for secure computation. We then present a comparison of six commonly available Homomorphic Encryption libraries - SEAL, HElib, TFHE, Paillier, ELGamal and RSA across these identified features. Support for different languages and real-life applications are also elucidated.
Homomorphic encryption is a method used in cryptopgraphy to create programs that can interact with encrypted data without ever leaving the data in the clear. This has many potential applications in cybersecurity. This paper uses automorphisms of affine space to create a form of homomorphic encryption for straight line programs. The encryption method used for the data is known as multivariate encryption. This gives a potentially powerful new method in cyber security based on techniques from algebraic geometry.
Georgia State University（佐治亚州立大学）；Rochester Institute of Technology（罗切斯特理工学院）
Outsourced computation for neural networks allows users access to state of the art models without needing to invest in specialized hardware and know-how. The problem is that the users lose control over potentially privacy sensitive data. With homomorphic encryption (HE) computation can be performed on encrypted data without revealing its content. In this systematization of knowledge, we take an in-depth look at approaches that combine neural networks with HE for privacy preservation. We categorize the changes to neural network models and architectures to make them computable over HE and how th…
This paper proposes a new homomorphic functional encryption using modular multiplications over a hidden ring. Unlike traditional homomorphic encryption where users can only passively perform ciphertext addition or multiplication, the homomorphic functional encryption retains homomorphic addition and scalar multiplication properties, but also allows for the user's inputs through polynomial variables. The proposed homomorphic encryption can be applied to any polynomials over a finite field, with their coefficients considered as their privacy. We denote the polynomials before homomorphic encrypti…
Homomorphic encryption and blockchain technology are regarded as two significant technologies for improving e-voting systems. In this paper, we suggest a novel e-voting system using homomorphic encryption and blockchain technology that is focused on encrypting voter data. By encrypting voter information rather than cast votes, the system enables various statistical analyses regarding the vote result while securing the credibility, privacy, and verifiability of overall e-voting system. We checked the efficiency of the overall system by comparing the speed of the proposed system with the speed o…
Quadratic time $O(n^2)$ Fully Homomorphic public key encryption algorithm based on a unique technique to create p-adic homomorphism from ring X to ring Y
A database is a prime target for cyber-attacks as it contains confidential, sensitive, or protected information. With the increasing sophistication of the internet and dependencies on internet data transmission, it has become vital to be aware of various encryption technologies and trends. It can assist in safeguarding private information and sensitive data, as well as improve the security of client-server communication. Database encryption is a procedure that employs an algorithm to convert data contained in a database into "cipher text," which is incomprehensible until decoded. Homomorphic e…
We propose a symmetric key homomorphic encryption scheme based on the evaluation of multivariate polynomials over a finite field. The proposed scheme is somewhat homomorphic with respect to addition and multiplication. Further, we define a generalization of the Learning with Errors problem called the Hidden Subspace Membership problem and show that the semantic security of the proposed scheme can be reduced to the hardness of this problem.
Institute for Development and Research in Banking Technology（银行技术发展与研究所）；University of Hyderabad（海得拉巴大学）
The Machine Learning and Deep Learning Models require a lot of data for the training process, and in some scenarios, there might be some sensitive data, such as customer information involved, which the organizations might be hesitant to outsource for model building. Some of the privacy-preserving techniques such as Differential Privacy, Homomorphic Encryption, and Secure Multi-Party Computation can be integrated with different Machine Learning and Deep Learning algorithms to provide security to the data as well as the model. In this paper, we propose a Chaotic Extreme Learning Machine and its …
This is a survey on some topics in Lattice based cryptography and Homomorphic Encryption. In particular, we define some lattice problems, LWE and RLWE, and state the reductions given by Regev and Peikert. We also give a full treatment of the recent CKKS homomorphic encryption scheme and give some worked out examples.
A Generative Adversarial Network (GAN) is a deep-learning generative model in the field of Machine Learning (ML) that involves training two Neural Networks (NN) using a sizable data set. In certain fields, such as medicine, the training data may be hospital patient records that are stored across different hospitals. The classic centralized approach would involve sending the data to a centralized server where the model would be trained. However, that would involve breaching the privacy and confidentiality of the patients and their data, which would be unacceptable. Therefore, Federated Learning…
The breakthrough of achieving fully homomorphic encryption sparked enormous studies on where and how to apply homomorphic encryption schemes so that operations can be performed on encrypted data without the secret key while still obtaining the correct outputs. Due to the computational cost, inflated ciphertext size and limited arithmetic operations that are allowed in most encryption schemes, feasible applications of homomorphic encryption are few. While theorists are working on the mathematical and cryptographical foundations of homomorphic encryption in order to overcome the current limitati…
Hong Kong University of Science and Technology（香港科技大学）；Clustar（星辰天合）；Peng Cheng Lab（鹏城实验室）
With the increasing awareness of privacy protection and data fragmentation problem, federated learning has been emerging as a new paradigm of machine learning. Federated learning tends to utilize various privacy preserving mechanisms to protect the transferred intermediate data, among which homomorphic encryption strikes a balance between security and ease of utilization. However, the complicated operations and large operands impose significant overhead on federated learning. Maintaining accuracy and security more efficiently has been a key problem of federated learning. In this work, we inves…
École Supérieure en Informatique, Sidi Bel Abbès（西迪贝勒阿巴斯高等信息学院）；OpenMined
Machine learning algorithms have achieved remarkable results and are widely applied in a variety of domains. These algorithms often rely on sensitive and private data such as medical and financial records. Therefore, it is vital to draw further attention regarding privacy threats and corresponding defensive techniques applied to machine learning models. In this paper, we present TenSEAL, an open-source library for Privacy-Preserving Machine Learning using Homomorphic Encryption that can be easily integrated within popular machine learning frameworks. We benchmark our implementation using MNIST…
We consider an architecture of confidential cloud-based control synthesis based on Homomorphic Encryption (HE). Our study is motivated by the recent surge of data-driven control such as deep reinforcement learning, whose heavy computational requirements often necessitate an outsourcing to the third party server. To achieve more flexibility than Partially Homomorphic Encryption (PHE) and less computational overhead than Fully Homomorphic Encryption (FHE), we consider a Reinforcement Learning (RL) architecture over Leveled Homomorphic Encryption (LHE). We first show that the impact of the encryp…
Deep Learning as a Service (DLaaS) stands as a promising solution for cloud-based inference applications. In this setting, the cloud has a pre-learned model whereas the user has samples on which she wants to run the model. The biggest concern with DLaaS is user privacy if the input samples are sensitive data. We provide here an efficient privacy-preserving system by employing high-end technologies such as Fully Homomorphic Encryption (FHE), Convolutional Neural Networks (CNNs) and Graphics Processing Units (GPUs). FHE, with its widely-known feature of computing on encrypted data, empowers a wi…
Homomorphic encryption has been an area of study in classical computing for decades. The fundamental goal of homomorphic encryption is to enable (untrusted) Oscar to perform a computation for Alice without Oscar knowing the input to the computation or the output from the computation. Alice encrypts the input before sending it to Oscar, and Oscar performs the computation directly on the encrypted data, producing an encrypted result. Oscar then sends the encrypted result of the computation back to Alice, who can decrypt it. We describe an approach to homomorphic encryption for quantum annealing …
Recently, deep learning as a service (DLaaS) has emerged as a promising way to facilitate the employment of deep neural networks (DNNs) for various purposes. However, using DLaaS also causes potential privacy leakage from both clients and cloud servers. This privacy issue has fueled the research interests on the privacy-preserving inference of DNN models in the cloud service. In this paper, we present a practical solution named BAYHENN for secure DNN inference. It can protect both the client's privacy and server's privacy at the same time. The key strategy of our solution is to combine homomor…
Fully homomorphic encryption (FHE) protects data privacy in cloud computing by enabling computations to directly occur on ciphertexts. To improve the time-consuming FHE operations, we present an electro-optical (EO) FHE accelerator, CryptoLight. Compared to prior FHE accelerators, on average, CryptoLight reduces the latency of various FHE applications by >94.4% and the energy consumption by >95%
University of Stavanger（斯塔万格大学）；Virginia Commonwealth University（弗吉尼亚联邦大学）；Old Dominion University（奥尔多明尼昂大学）；Norwegian University of Science and Technology（挪威科技大学）
Medical data is often highly sensitive in terms of data privacy and security concerns. Federated learning, one type of machine learning techniques, has been started to use for the improvement of the privacy and security of medical data. In the federated learning, the training data is distributed across multiple machines, and the learning process is performed in a collaborative manner. There are several privacy attacks on deep learning (DL) models to get the sensitive information by attackers. Therefore, the DL model itself should be protected from the adversarial attack, especially for applica…
Fully Homomorphic Encryption (FHE) emerges one of the most promising solutions to privacy-preserving computing in an untrusted cloud. FHE can be implemented by various schemes, each of which has distinctive advantages, i.e., some are good at arithmetic operations, while others are efficient when implementing Boolean logic operations. Therefore, it is difficult for even cryptography experts let alone average users to choose the "right" FHE scheme to efficiently implement a specific application. Prior work only qualitatively compares few FHE schemes. In this paper, we present an empirical study,…
