International Association
for Cryptologic Research

Event date: 16 April to 20 April 2018
Submission deadline: 12 March 2018
Notification: 15 March 2018

Event date: 28 December to 30 December 2018
Submission deadline: 25 June 2018
Notification: 6 August 2018

Logistic regression is a popular technique used in machine learning to construct classification models. Since the construction of such models is based on computing with large datasets, it is an appealing idea to outsource this computation to a cloud service. The privacy-sensitive nature of the input data requires appropriate privacy preserving measures before outsourcing it. Homomorphic encryption enables one to compute on encrypted data directly, without decryption, and can be used to mitigate the privacy concerns raised by using a cloud service. In this paper, we propose an algorithm (and its implementation) to train a logistic regression model on a homomorphically encrypted dataset. The core of our algorithm consists of a new iterative method that can be seen as a simplified form of the fixed Hessian method, but with a much lower multiplicative complexity. We test the new method on two interesting real life applications: the first application is in medicine and constructs a model to predict the probability for a patient to have cancer, given genomic data as input; the second application is in finance and the model predicts the probability of a credit card transaction to be fraudulent. The method produces accurate results for both applications, comparable to running standard algorithms on plaintext data. This article introduces a new simple iterative algorithm to train a logistic regression model that is tailored to be applied on a homomorphically encrypted dataset. This algorithm can be used as a privacy-preserving technique to build a binary classification model and can be applied in a wide range of problems that can be modelled with logistic regression. Our implementation results show that our method can handle the large datasets used in logistic regression training.

A cryptosystem which supports both addition and multiplication (thereby preserving the ring structure of the plaintexts) is known as fully homomorphic encryption (FHE) and is very powerful. Using such a scheme, any circuit can be homomorphically evaluated, effectively allowing the construction of programs which may be run on ciphertexts of their inputs to produce a ciphertext of their output. Since such a program never decrypts its input, it can be run by an untrusted party without revealing its inputs and internal state. The existence of an efficient and fully homomorphic cryptosystem would have great practical implications in the outsourcing of private computations, for instance, in the context of cloud computing. In previous work I proposed the fully homomorphic public-key encryption scheme with the size of ciphertext which is not small enough. In this paper the size of ciphertext is one-eighth of the size in the previously proposed scheme. Because proposed scheme adopts the medium text with zero norm, it is immune from the the “p and -p attack”. As the proposed scheme is based on computational difficulty to solve the multivariate algebraic equations of high degree, it is immune from the Gröbner basis attack, the differential attack, rank attack and so on.

Quantum entanglement is of great importance to quantum cryptography and computation. So far, all experimental demonstrations of entanglement are designed to check Bell's inequality which is based on Bell's formulation for EPR paradox. In this note, we specify the assumptions needed in Bell's mathematical argument. We then show the contradictions among these assumptions. As a result, it becomes very easy to see that Bell's inequality is trivial.

In this paper, we introduce Saber, a package of cryptographic primitives whose security relies on the hardness of the Module Learning With Rounding problem (Mod-LWR). We first describe a secure Diffie-Hellman type key exchange protocol, which is then transformed into an IND-CPA encryption scheme and finally into an IND-CCA secure key encapsulation mechanism using a post-quantum version of the Fujisaki-Okamoto transform. The design goals of this package were simplicity, efficiency and flexibility resulting in the following choices: all integer moduli are powers of $2$ avoiding modular reduction and rejection sampling entirely; the use of LWR halves the amount of randomness required compared to LWE-based schemes and reduces bandwidth; the module structure provides flexibility by reusing one core component for multiple security levels. A constant-time AVX2 optimized software implementation of the KEM with parameters providing more than 128 bits of post-quantum security, requires only 101K, 125K and 129K cycles for key generation, encapsulation and decapsulation respectively on a Dell laptop with an Intel i7-Haswell processor.

NTRUEncrypt is one of the most promising candidates for quantum-safe cryptography. In this paper, we focus on the NTRU743 paramter set. We give a report on all known attacks against this parameter set and show that it delivers 256 bits of security against classical attackers and 128 bits of security against quantum attackers. We then present a parameter-dependent optimization using a tailored hierarchy of multipli- cation algorithms as well as the Intel AVX2 instructions, and show that this optimization is constant-time. Our implementation is two to three times faster than the reference implementation of NTRUEncrypt.

While non-interactive zero-knowledge (NIZK) proofs require trusted parameters, Groth, Ostrovsky and Sahai constructed non-interactive witness-indistinguishable (NIWI) proofs without any setup; they called their scheme a non-interactive zap. More recently, Bellare, Fuchsbauer and Scafuro investigated the security of NIZK in the face of parameter subversion and observe that NI zaps provide subversion-resistant soundness and WI. Arguments of knowledge prove that not only the statement is true, but also that the prover knows a witness for it, which is essential for anonymous identification. We present the first NIWI argument of knowledge without parameters, i.e., a NI zap of knowledge. Consequently, our scheme is also the first subversion-resistant knowledge-sound proof system, a notion recently proposed by Fuchsbauer.

It is well-known that non-comparison-based techniques can allow us to sort $n$ elements in $o(n \log n)$ time on a Random-Access Machine (RAM). On the other hand, it is a long-standing open question whether (non-comparison-based) circuits can sort $n$ elements from the domain $[1..2^k]$ with $o(k n \log n)$ boolean gates. We consider weakened forms of this question: first, we consider a restricted class of sorting where the number of distinct keys is much smaller than the input length; and second, we explore Oblivious RAMs and probabilistic circuit families, i.e., computational models that are somewhat more powerful than circuits but much weaker than RAM. We show that Oblivious RAMs and probabilistic circuit families can sort $o(\log n)$-bit keys in $o(n \log n)$ time or $o(k n \log n)$ circuit complexity where $n$ is the input length. We also show that in the balls-and-bins model of sorting where each key may carry an opaque ball that can only be moved around atomically but cannot be computed upon, our result achieves optimality, in that any oblivious algorithm or probabilistic circuit family that sorts $n$ balls each with a $\Omega(\log n)$-bit key must incur at least $\Omega(n \log n)$ atomic movement operations on balls. We extend our result to support the case when the keys are chosen from a large space but the number of distinct keys is small.

Finally, we optimize the IO efficiency of our oblivious algorithms for RAMs --- we show that even the $1$-bit special case of our algorithm can solve open questions regarding whether there exist oblivious algorithms for tight compaction and selection in linear IO.

The random-permutation model (RPM) and the ideal-cipher model (ICM) are idealized models that offer a simple and intuitive way to assess the conjectured standard-model security of many important symmetric-key and hash-function constructions. Similarly, the generic-group model (GGM) captures generic algorithms against assumptions in cyclic groups by modeling encodings of group elements as random injections and allows to derive simple bounds on the advantage of such algorithms.

Unfortunately, both well-known attacks, e.g., based on rainbow tables (Hellman, IEEE Transactions on Information Theory '80), and more recent ones, e.g., against the discrete-logarithm problem (Corrigan-Gibbs and Kogan, EUROCRYPT '18), suggest that the concrete security bounds one obtains from such idealized proofs are often completely inaccurate if one considers non-uniform or preprocessing attacks in the standard model. To remedy this situation, this work

1) defines the auxiliary-input (AI) RPM/ICM/GGM, which capture both non-uniform and preprocessing attacks by allowing an attacker to leak an arbitrary (bounded-output) function of the oracle's function table;

2) derives the first non-uniform bounds for a number of important practical applications in the AI-RPM/ICM, including constructions based on the Merkle-Damgard and sponge paradigms, which underly the SHA hashing standards, and for AI-RPM/ICM applications with computational security; and

3) using simpler proofs, recovers the AI-GGM security bounds obtained by Corrigan-Gibbs and Kogan against preprocessing attackers, for a number of assumptions related to cyclic groups, such as discrete logarithms and Diffie-Hellman problems, and provides new bounds for two assumptions.

An important step in obtaining these results is to port the tools used in recent work by Coretti et al. (EUROCRYPT '18) from the ROM to the RPM/ICM/GGM, resulting in very powerful and easy-to-use tools for proving security bounds against non-uniform and preprocessing attacks.

Many voting systems rely on art, rather than science, to ensure that votes are freely made, with equal influence. Such systems build upon creativity and skill, rather than scientific foundations. These systems are routinely broken in ways that compromise free-choice or permit undue influence. Breaks can be avoided by proving that voting systems satisfy formal notions of voters voting freely and of detecting undue influence. This manuscript provides a detailed technical introduction to a definition of ballot secrecy by Smyth that formalises the former notion and a definition of verifiability by Smyth, Frink & Clarkson that formalises the latter. The definitions are presented in the computational model of cryptography: Ballot secrecy is expressed as the inability to distinguish between an instance of the voting system in which voters cast some votes, from another instance in which the voters cast a permutation of those votes. Verifiability decomposes into individual verifiability, which is expressed as the inability to cause a collision between ballots, and universal verifiability, which is expressed as the inability to cause an incorrect election outcome to be accepted. The definitions are complimented with simple examples that demonstrate the essence of these properties and detailed proofs are constructed to show how secrecy and verifiability can be formally proved. Finally, the Helios and Helios Mixnet voting systems are presented as case studies to provide an understanding of state-of-the-art systems that are being used for binding elections.

We present a semantically secure somewhat homomorphic public-key cryptosystem working in sub-groups of $\mathbb{Z}_{n}^{*}$ of prime power order. Our scheme introduces a novel threshold homomorphic property, which we use to build a two-party protocol for secure integer comparison. In contrast to related work which encrypts and acts on each bit of the input separately, our protocol compares multiple input bits simultaneously within a single ciphertext. Compared to the related protocol of Damg\r{a}rd et al.~we present results showing this approach to be both several times faster in computation and lower in communication complexity.

Password-authenticated key exchange (PAKE) protocols allow users sharing a password to agree upon a high entropy secret. In this paper, a provably secure password-authenticated pro- tocol for group key establishment in the common reference string (CRS) model is presented. Our protocol is quite ecient, as regardless of the number of involved participants it can be imple- mented with only three communication rounds. We use a (by now classical) trick of Burmester and Desmedt for deriving group key exchange protocols using a two-party construction as main building block. In our case, the two party PAKE used as a base is a one-round protocol by Katz and Vaikuntanatan, which in turn builds upon a special kind of smooth projective hash functions (KV-SPHFs). As evidenced by Benhamouda et al., KV-SPHFs can be instantiated on Cramer-Shoup ciphertexts, thus yielding very ecient (and pairing free) constructions.

Event date: 24 September to 26 September 2018
Submission deadline: 30 April 2018
Notification: 18 June 2018

Event date: 19 August 2018
Submission deadline: 1 June 2018

A recent paper by Derler, Ramacher, and Slamanig (IEEE EuroS&P 2018) constructs double-authentication preventing signatures ("DAP signatures", a specific self-enforcement enabled variant of signatures where messages consist of an address and a payload) that have---if the supported address space is not too large---keys and signatures that are considerably more compact than those of prior work. We embark on their approach to restrict attention to small address spaces and construct novel DAP schemes that beat their verification key sizes by a factor of two, their signature sizes by a factor of five, and reduce their signing key sizes from linear to constant. We construct our DAP signatures generically from identification protocols, using a transform similar to but crucially different from that of Fiat and Shamir. We use random oracles. We don't use pairings.

Some voting systems are reliant on external authentication services. Others use cryptography to implement their own. We combine digital signatures and non-interactive proofs to derive a generic construction for voting systems with their own authentication mechanisms, from systems that rely on external authentication services. We prove that our construction produces systems satisfying ballot secrecy and election ver- ifiability, assuming the underlying voting system does. Moreover, we observe that works based on similar ideas provide neither ballot secrecy nor election verifiability. Finally, we demonstrate applicability of our results by applying our construction to the Helios voting system.

Memory Hard Functions (MHFs) have been proposed as an answer to the growing inequality between the computational speed of general purpose CPUs and Application Specific Integrated Circuits (ASICs). MHFs have seen widespread applications including password hashing, key stretching and proofs of work. Several metrics have been proposed to quantify the `memory hardness' of a function. Cumulative memory complexity (CMC) (Alwen and Serbinenko, STOC 2015) (or amortized Area $\times$ Time complexity (Alwen et. al., CCS 2017)) attempts to quantify the amortized cost to acquire/build the hardware to evaluate the function --- amortized by the number of instances of the function that can be evaluated of this hardware. By contrast, bandwidth hardness (Ren and Devadas, TCC 2017) attempts to quantify the amortized energy costs of evaluating this function on hardware --- which in turn is largely dominated by the number of cache misses. Ideally, a good MHF would be both bandwidth hard and have high cumulative memory complexity. While the cumulative memory complexity of leading MHF candidates is well understood, little is known about the bandwidth hardness of many of the most prominent MHF candidates.

Our contributions are as follows: First, we provide the first reduction proving that, in the parallel random oracle model, the bandwidth hardness of a Data-Independent Memory Hard Function (iMHF) is described by the red-blue pebbling cost of the directed acyclic graph (DAG) associated with that iMHF. Second, we show that the goals of designing an MHF with high CMC/bandwidth hardness are well aligned. In particular, we prove that any function with high CMC also has relatively high bandwidth costs. This result leads to the first unconditional lower bound on the bandwidth cost of scrypt. Third, we analyze the bandwidth hardness of several prominent iMHF candidates such as Argon2i (Biryukov et. al., 2015), winner of the password hashing competition, aATSample and DRSample (Alwen et. al., CCS 2017) --- the first practical iMHF with asymptotically optimal CMC. More specifically, we show that Argon2i is maximally bandwidth hard as long as the cache-size $m$ is at most $m \in O(n^{2/3-\epsilon})$ where $n$ is the total number of data-labels produced during computation. We also show that aATSample and DRSample are maximally bandwidth hard as long as the cache-size is $m \in O(n^{1-\epsilon})$. Finally, we show that the problem of finding a red-blue pebbling with minimum bandwidth cost is NP-hard.

Oblivious RAM is a well-known cryptographic primitive to hide data access patterns. However, the best known ORAM schemes require a logarithmic computation time in the general case which makes it infeasible for use in real-world applications. In practice, hiding data access patterns should incur a constant latency per access.

In this work, we present PRO-ORAM --- an ORAM construction that achieves constant latencies per access in a large class of applications. PRO-ORAM theoretically and empirically guarantees this for read-only data access patterns, wherein data is written once followed by read requests. It makes hiding data access pattern practical for read-only workloads, incurring sub-second computational latencies per access for data blocks of 256 KB, over large (gigabyte-sized) datasets.PRO-ORAM supports throughputs of tens to hundreds of MBps for fetching blocks, which exceeds network bandwidth available to average users today. Our experiments suggest that dominant factor in latency offered by PRO-ORAM is the inherent network throughput of transferring final blocks, rather than the computational latencies of the protocol. At its heart, PRO-ORAM utilizes key observations enabling an aggressively parallelized algorithm of an ORAM construction and a permutation operation, as well as the use of trusted computing technique (SGX) that not only provides safety but also offers the advantage of lowering communication costs.

Lightweight block ciphers rely on simple operations to allow compact implementation. Thanks to its efficiency, bit permutation has emerged as an optimal choice for state-wise diffusion. It can be implemented by simple wiring or shifts. However, as recently shown by Spectre and Meltdown attacks, efficiency and security often go against each other.

In this work, we show how bit permutations introduce a side-channel vulnerability that can be exploited to extract the secret key from the cipher. Such vulnerabilities are specific to bit permutations and do not occur in other state-wise diffusion alternatives. We propose Side-Channel Assisted Differential-Plaintext Attack (SCADPA) which targets this vulnerability in bit permutation operation. SCADPA is experimentally demonstrated on PRESENT-80 on an 8-bit microcontroller, with the best case key recovery in 17 encryptions. The attack is then extended to latest bit-permutation based cipher GIFT, allowing full key recovery in 36 encryptions. We also propose and experimentally verify an automatic threshold method which can be easily applied to SCADPA, allowing automation of the attack. Moreover, SCADPA on bit permutations has other applications. Application for reverse engineering secret sboxes in PRESENT-like proprietary ciphers is shown. We also highlight a special case, where fixing one vulnerability opens another one. This is shown by applying SCADPA on some assembly level fault attack countermeasures, rendering it less secure than unprotected implementations. Lastly, we also provide several different attack scenarios, such as targeting different encryption modes.