With the advent of cloud computing, traditional machine learning (ML) are migrating into cloud-based ML day by day following the concept of machine learning as a cloud service, which enables multiple entities to contribute to and benefit from shared datasets and models. As well as training the linear classification model, training the nonlinear classification model is also an essential task in cloud-based ML. However, this task commonly involves learning knowledge from different datasets provided by various entities, which often contain sensitive information like patients' physiological indices. Therefore, it gives rise a natural question how to allow multiple users collaboratively participating in a nonlinear classification task while preserving these datas' privacy. As a promising cryptographic tool, the concept of unbounded multiinput functional encryption can be developed to answer such a question, such as google search engines are running over this concept-based ML approaches. However, most of existing approaches are derived from this concept with inner product functionality, specifying for a linear classification model and thus fails to cope with a non-linear classification one. In this paper, we introduce an advanced cryptographic concept called unbounded multi-input quadratic functional encryption, and give a concrete construction which allows arbitrary number of users participating in the classifying tasks with a nonlinear classification model but without divulging their private data. Moreover, we provide a strict mathematical security proof under a well-defined security model as well as some security attacks are analyzed, followed by an experimental analysis and comparison on a real dateset as well as a practical use case to demonstrate our scheme's performance.

Pseudorandom functions (PRFs) are a very important tool in cryptography, and the learning with rounding (LWR) problem is one of the main issues in their construction. LWR problem, is to find from ⌊ A s ⌋ p , where and is the rounding function. The LWR problem is considered a variant of the learning with error (LWE) problem, that is, to find s from b = A s + e , where , and LWE has been reduced to GapSVP and SIVP. The hardness of the lattice problems is the security foundation of the issued schemes. The best‐known reduction for LWR was completed using information‐theoretic entropy arguments, and the reduction requires q ≥ 2 n m p . It does not directly reduce to the closest vector problem (CVP) problem, but rather to the LWE problem. However, the reduction in the aforementioned work significantly reduces the difficulty of LWR. To more accurately characterize the hardness of LWR, this paper uses statistical approximation and a Quantum Fourier Transform to reduce LWR to the CVP, thereby ensuring the hardness of LWR. Furthermore, unlike the previous conclusions, our reduction involves minimal loss and has broad security conditions, requiring only that , where q and p are prime numbers and 0 < α < 1.

An identity-based matchmaking encryption (IB-ME) scheme proposed at JOC 2021 supports anonymous but authenticated communications in a way that communication parties can both specify the senders or receivers on the fly. IB-ME is easy to be used in several network applications requiring privacy-preserving for its efficient implementation and special syntax. Despite the rigorous security proofs in previous security models, the existing IB-ME schemes are still possibly vulnerable to some potential neglected attacks. Aiming at the above problems, we provide a stronger security definition of authenticity considering new attacks to fit real-world scenarios and then propose a generic construction of IB-ME satisfying the new model. Inspired by the prior IB-ME construction of Chen et al., the proposed scheme is constructed by combining 2-level anonymous hierarchical IBE (HIBE) and identity-based signature (IBS) schemes. In order to upgrade lattice-based IB-ME with better efficiency, we additionally improve a lattice IBS, as an independent technical contribution, to shorten its signature and thus reduce the final IB-ME ciphertext size. By combining the improved IBS and any 2-level adaptively-secure lattice-based HIBE with anonymity, we finally obtain the first lattice-based IB-ME construction achieving privacy and new-proposed stronger authenticity simultaneously.

As a common metric, maximum similarity between two objects is widely employed by web platforms to provide matching services. However, the calculation of maximum similarity involves numerous sensitive or confidential users’ data, and the web platform server is often not trusted who might peep these data out of curiosity, or even worse sell them to unauthorized entities to make profits. Therefore, many research lines on functional encryption have been suggested and studied on how to calculate the maximum similarity while ensure the privacy of users’ data. Unfortunately, all of them will divulge some intermediate results to the web platform server when processing this issue. In this paper we present a new functional encryption scheme supporting privacy-preserving maximum similarity, which enables the web service platforms to figure out the maximum similarity without learning anything else about their data. Moreover, we provide a formal analysis to prove the security of the proposed scheme, followed by some experimental evaluations and comprehensive comparisons with the related works. It shows that, our scheme is the first functional encryption realization on maximum similarity without divulging the intermediate result and meanwhile achieve a higher security-function privacy, as well as a traditional data privacy.

Universal composability (UC) is a primary security flavor for designing oblivious transfer (OT) due to its advantage of arbitrary composition. However, the study of UC‐secure OT over lattices is still far behind compared with constructions over prequantum assumptions. Relying on the learning with errors (LWE) assumption, Quach proposes a dual‐mode encryption scheme (SCN’20) for deriving a two‐round OT whose security is provably UC‐secure in the common reference string (CRS) model. Due to its use of a randomized rounding function proposed by Benhamouda et al. (PKC’18), this OT can only be limited to transmitting single‐bit messages. Therefore, conducting trivial repetitions of Quach’s OT when transmitting multibit strings would be very costly. In this work, we put forward a modified dual‐mode encryption cryptosystem under the decisional LWE assumption, from which we can derive a UC‐secure string OT with both full‐fledged dual‐mode security and better efficiency on transmitting strings. The key technique we adopt is a key reconciliation scheme proposed by Jiang et al. (PKC’20), which is utilized to extend the single‐bit symmetric encryption key (produced by the aforementioned rounding function) to a multibit case. Through a comprehensive performance analysis, we demonstrate that our proposal can indeed strike a balance between security and efficiency.

Abstract Multi‐key fully homomorphic encryption (MKFHE) allows computations on ciphertexts encrypted by different users, which can be applied to implement secure multi‐party computing (MPC). The current NTRU‐based MKFHE has the following two drawbacks: One is that the relinearisation process during homomorphic evaluation is so complicated that the corresponding computation time is costly. The other is that a class of subfield attacks are proposed and affects the security of NTRU schemes over power‐of‐2 cyclotomic rings for large moduli q, especially for the NTRU‐based fully homomorphic encryption (FHE) schemes. In this work, an efficient MKFHE scheme is proposed over prime cyclotomic rings with fewer relinearisations, which seems a good choice because of its potential to resist a subfield attack. More specifically, the time of the relinearisation process is reduced by half in homomorphic evaluations by separating the homomorphic multiplication and the relinearisation process (implementing two homomorphic multiplication operations together before relinearisation), while in current NTRU‐type MKFHE schemes, these two processes are usually performed together. The error bound of the basic function components is re‐analysed over prime cyclotomic rings in the average case, which can be used in the error analysis of our scheme. We construct an efficient NTRU‐based single‐key FHE scheme and an efficient MKFHE scheme over prime cyclotomic rings through relinearisation and modulus‐switching techniques. The MKFHE scheme proposed has the on‐the‐fly property and has a tight ciphertext size compared with the GSW‐type and BGV‐type MKFHE schemes. An experiment shows that the homomorphic evaluation of the optimised single‐key FHE scheme proposed is 1.9 times faster than an efficient NTRU‐type MKFHE DHS16 proposed at DCC 2016.

As a basic and important primitive, hash proof system can be used to construct many cryptographic schemes and protocols. Therefore, it is significant to instantiate more efficient hash proof systems from various assumptions. Although there are many hash proof systems based on various classical assumptions, only a handful of efficient hash proof systems are known based on post-quantum assumptions. In this paper, we present several new hash proof systems based on the standard learning with errors (LWE) problem, which is at least as hard as standard worst-case lattice problems. Comparing with other existing constructions based on lattices, our main advantages are 2-fold: much simpler and more efficient. And our constructions can be easily extended to be identity-based ones and updatable ones. Throughout the paper, our main idea is to base hash proof systems on a new subset indistinguishability problem related to LWE, and employ the property of smooth parameter of q-ary orthogonal lattices to ensure smoothness.

Abstract Proxy re‐encryption (PRE) is a cryptographic primitive that allows a proxy to turn an Alice's ciphertext into a Bob's ciphertext on the same plaintext. At present, there are many different PRE schemes that have been proposed with different properties. However, all of them are based on the logarithm assumption and the large integer factorization assumption except for a bidirectional PRE scheme over lattices. In this paper, we construct the first multi‐use unidirectional PRE scheme based on lattices. In addition, the generation of the PRE key does not interact with two users, and the scheme can resist collusion attacks. Moreover, it is proved chosen plaintext attack secure in the standard model based on the Learning With Errors assumption. Finally, an identity‐based PRE is obtained from the basic construction. Copyright © 2015 John Wiley & Sons, Ltd.

Key-dependent message KDM security should be considered in the design of security protocols, especially for complicated ones, where the messages related to the secret key might be encrypted. In this paper, we present a new method of constructing a KDM secure asymmetric encryption scheme with the notation of hybrid encryption in the standard model; although the notation of hybrid encryption was thought as no help to get rid of dependencies between messages and the secret key. Our result can also be seen as a partial instantiation for a previously well-known KDM secure asymmetric encryption scheme based on random oracle. As we know, this has never been carried out before. And our result indicates a new cryptographic application for the primitive of lossy trapdoor function. Throughout the paper, our main idea is to archive KDM security by making use of both leakage-resilience and auxiliary-input security properties. Copyright © 2014 John Wiley & Sons, Ltd.