Fully Homomorphic Encryption (FHE) in the realm of deep neural network (DNN) encrypted inference represents a pivotal advancement in privacy-preserving machine learning. This technology allows users to securely access DNN inference services hosted on remote servers without compromising their personal privacy. Given its wide range of potential applications, FHE has garnered significant research attention. Despite its rapid progress, FHE still faces considerable challenges, particularly the high computational resource demands. Moreover, varying configurations of FHE schemes can lead to notable differences in performance, whether in terms of efficiency or inference accuracy, making it difficult to strike an optimal balance tailored to specific application requirements.To tackle this challenge, we introduce a new approach that simulates FHE-induced errors to assess the impact of different FHE architectures and parameter configurations on encrypted inference during the model testing phase. Our simulation framework enables efficient approximation of homomorphic computation outcomes on a DNN model, specifically for third-generation FHEs, without the need for executing the complete homomorphic process. This method significantly streamlines the research process for optimizing parameters based on specific application needs. We validate the effectiveness of our approach through extensive performance benchmarking across a variety of experimental settings.

Third-generation Fully Homomorphic Encryption (FHE), particularly the FHEW/TFHE schemes, is recognized for its balanced security requirements, small parameters, and low memory usage, though the current methods in the scenarios of Deep Neural Network (DNN) inference still have high computational costs, limiting the practical applicability. This work demonstrates how to improve practicality of the third-generation technologies for DNN tasks while preserving its key advantages. Our work focuses on two main contributions. First, we developed a computational architecture called FHE-Neuron, which reconfigures the parameters and bootstrapping structure of traditional FHEW/TFHE Boolean operations. This architecture significantly reducing the cost of encrypted DNN inference by dynamically switching the precision of encrypted data during computation—using high precision for cost-effective linear operations and low precision for computationally expensive nonlinear operations. Second, we introduced an FHE-aware Quantization and Fine-tuning framework that optimizes model parameters to align with FHE-Neuron’s constraints, ensuring high accuracy in encrypted inference. We validate our approach on various neural network models across several computing platforms. In our experiments, our method achieves one-image inference time on average 4.5 milliseconds for MNIST and 17 milliseconds for Fashion MNIST, achieving accuracy rates of 96.52% and 88.57% respectively. For the CIFAR-10 dataset, our system completes one image inference in 30 seconds with a 90.5% accuracy rate.

This work introduces KsPIR, a new practically efficient single-server private information retrieval (PIR) system that outperforms the state-of-the-art Spiral (Menon and Wu, S&P 2022) in terms of server response times. We achieve this by proposing novel dimension folding methods, inspired by recent advancements in fully homomorphic encryption. Our methods offer two significant advantages: firstly, they feature simpler designs that eliminate the need for ciphertext expansion steps in Spiral. Secondly, and more importantly, we propose two types of designs that offer distinct advantages - the first type enables preprocessing of the most resource-intensive computation in the offline stage before receiving the query, thereby optimizing online response time; the second type optimizes overall response time without requiring preprocessing in the offline stage, accomplished through a highly optimized baby-step-giant-step matrix-vector homomorphic multiplication.

Fully Homomorphic Encryption (FHE) is vital for computing over encrypted data, thereby enabling numerous privacy-preserving applications. This work focuses on the third generation FHE schemes (e.g., FHEW and TFHE), known for their fast bootstrapping, small FHE parameters, and robust security built on milder assumptions.

Abstract Zero-knowledge succinct non-interactive arguments of knowledge (zk-SNARKs) are cryptographic protocols that offer efficient and privacy-preserving means of verifying NP language relations and have drawn considerable attention for their appealing applications, e.g., verifiable computation and anonymous payment protocol. Compared with the pre-quantum case, the practicability of this primitive in the post-quantum setting is still unsatisfactory, especially for the space complexity. To tackle this issue, this work seeks to enhance the efficiency and compactness of lattice-based zk-SNARKs, including proof length and common reference string (CRS) length. In this paper, we develop the framework of square span program-based SNARKs and design new zk-SNARKs over cyclotomic rings. Compared with previous works, our construction is without parallel repetition and achieves shorter proof and CRS lengths than previous lattice-based zk-SNARK schemes. Particularly, the proof length of our scheme is around $$23.3\%$$ 23.3 % smaller than the recent shortest lattice-based zk-SNARKs by Ishai et al. (in: Proceedings of the 2021 ACM SIGSAC conference on computer and communications security, pp 212–234, 2021), and the CRS length is $$3.6\times$$ 3.6 × smaller. Our constructions follow the framework of Gennaro et al. (in: Proceedings of the 2018 ACM SIGSAC conference on computer and communications security, pp 556–573, 2018), and adapt it to the ring setting by slightly modifying the knowledge assumptions. We develop concretely small constructions by using module-switching and key-switching procedures in a novel way.

Abstract Non-malleable codes were introduced by Dziembowski et al. (in: Yao (ed) ICS2010, Tsinghua University Press, 2010), and its main application is the protection of cryptographic devices against tampering attacks on memory. In this work, we initiate a comprehensive study on non-malleable codes for the class of partial functions, that read/write on an arbitrary subset of codeword bits with specific cardinality. We present two constructions: the first one is in the CRS model and allows the adversary to selectively choose the subset of codeword bits, while the latter is in the standard model and adaptively secure. Our constructions are efficient in terms of information rate, while allowing the attacker to access asymptotically almost the entire codeword. In addition, they satisfy a notion which is stronger than non-malleability, that we call non-malleability with manipulation detection, guaranteeing that any modified codeword decodes to either the original message or to $$\bot $$ ⊥ . We show that our primitive implies All-Or-Nothing Transforms (AONTs), and as a result our constructions yield efficient AONTs under standard assumptions (only one-way functions), which, to the best of our knowledge, was an open question until now. Furthermore, we construct a notion of continuous non-malleable codes (CNMC), namely CNMC with light updates, that avoids the full re-encoding process and only uses shuffling and refreshing operations. Finally, we present a number of additional applications of our primitive in tamper resilience.

Threshold private set intersection (TPSI) allows multiple parties to learn the intersection of their input sets only if the size of the intersection is greater than a certain threshold. This task has been demonstrated useful with practical applications, and thus many active research has been conducted. However, current solutions for TPSI are still slow for large input sets e.g., n=2^20 for the set size, and the potentially practical candidates are only secure against semi-honest adversaries. For the basic PSI, there have been efficient and scalable solutions, even in the malicious settings. It is interesting to determine whether adding a threshold feature would inherently incur a large overhead to PSI.

In this work, we significantly improve the efficiency of non-malleable codes in the split state model, by constructing a code with codeword length (roughly), where |s| is the length of the message, and k is the security parameter. This is a substantial improvement over previous constructions, both asymptotically and concretely.

The smoothing parameter ηε(L) of a Euclidean lattice L, introduced by Micciancio and Regev (FOCS'04; SICOMP'07), is (informally) the smallest amount of Gaussian noise that “smooths out” the discrete structure of L (up to error ε). It plays a central role in the best known worst-case/average-case reductions for lattice problems, a wealth of lattice-based cryptographic constructions, and (implicitly) the tightest known transference theorems for fundamental lattice quantities. In this work we initiate a study of the complexity of approximating the smoothing parameter to within a factor γ, denoted γ-GapSPP. We show that (for ε = 1/ poly(n)): . (2+o(1))-GapSPP ∈ AM, via a Gaussian analogue of the classic Goldreich-Goldwasser protocol (STOC'98); . (1 + o(1))-GapSPP ∈ coAM, via a careful application of the Goldwasser-Sipser (STOC'86) set size lower bound protocol to thin shells in Rn; . (2 + o(1))-GapSPP E SZK ⊆ AM ∩ coAM (where SZK is the class of problems having statistical zero-knowledge proofs), by constructing a suitable instance-dependent commitment scheme (for a slightly worse o(1)-term); . (1 + o(1))-GapSPP can be solved in deterministic 2O(n)polylog(1/ε) time and 2O(n)space. As an application, we demonstrate a tighter worst-case to average-case reduction for basing cryptography on the worstcase hardness of the GapSPP problem, with Õ(√n) smaller approximation factor than the GapSVP problem. Central to our results are two novel, and nearly tight, characterizations of the magnitude of discrete Gaussian sums over L: the first relates these directly to the Gaussian measure of the Voronoi cell of L, and the second to the fraction of overlap between Euclidean balls centered around points of L.