Research Article
Maryam Azadmanesh; Behrouz Shahgholi Ghahfarokhi; Maede Ashouri-Talouki
Abstract
Using generative models to produce unlimited synthetic samples is a popular replacement for database sharing. Generative Adversarial Network (GAN) is a popular class of generative models which generates synthetic data samples very similar to real training datasets. However, GAN models do not necessarily ...
Read More
Using generative models to produce unlimited synthetic samples is a popular replacement for database sharing. Generative Adversarial Network (GAN) is a popular class of generative models which generates synthetic data samples very similar to real training datasets. However, GAN models do not necessarily guarantee training privacy as these models may memorize details of training data samples. When these models are built using sensitive data, the developers should ensure that the training dataset is appropriately protected against privacy leakage. Hence, quantifying the privacy risk of these models is essential. To this end, this paper focuses on evaluating the privacy risk of publishing the generator network of GAN models. Specially, we conduct a novel generator white-box membership inference attack against GAN models that exploits accessible information about the victim model, i.e., the generator’s weights and synthetic samples, to conduct the attack. In the proposed attack, an auto-encoder is trained to determine member and non-member training records. This attack is applied to various kinds of GANs. We evaluate our attack accuracy with respect to various model types and training configurations. The results demonstrate the superior performance of the proposed attack on non-private GANs compared to previous attacks in white-box generator access. The accuracy of the proposed attack is 19% higher on average than similar work. The proposed attack, like previous attacks, has better performance for victim models that are trained with small training sets.
Research Article
Abstract
Digital signature schemes are used to guarantee for non-repudiation and authenticity of any kind of data like documents, messages or software. The Winternitz one-time signature (WOTS) scheme, which can be described using a certain number of so-called “function chains”, plays an important ...
Read More
Digital signature schemes are used to guarantee for non-repudiation and authenticity of any kind of data like documents, messages or software. The Winternitz one-time signature (WOTS) scheme, which can be described using a certain number of so-called “function chains”, plays an important role in the design of both stateless and stateful many-time signature schemes. The main idea of WOTS scheme is the use of a limited number of function chains, all of which begin at some random values. This work introduces WOTS-GES, a new WOTS type signature scheme in which the need for computing all of the intermediate values of the chains is eliminated. More precisely, to compute each algorithm of the proposed scheme, we only need to calculate one intermediate value. This significantly reduces the number of required operations needed to calculate the algorithms of WOTS-GES. To achieve this results, we have used the concept of “leveled” multilinear maps which is alsoreferred to as graded encoding schemes. We expect these results to increase the efficiency of Winternitz based digital signature schemes.
Research Article
Hamid Mala; Mohammad Reza Saeidi
Abstract
In the last two decades bilinear pairings have found many applications in cryptography. Meanwhile identity-based cryptosystems based on bilinear pairings have received particular attention. The IEEE, IETF, and ISO organizations have been working on standardization of pairing-based cryptographic schemes. ...
Read More
In the last two decades bilinear pairings have found many applications in cryptography. Meanwhile identity-based cryptosystems based on bilinear pairings have received particular attention. The IEEE, IETF, and ISO organizations have been working on standardization of pairing-based cryptographic schemes. The Boneh-Franklin identity-based encryption and Sakai-Kasahara identity-based signature are the most well-known identity-based schemes that have been standardized. So far, various schemes have been proposed to reduce the computational overhead of pairing operations. All these schemes are trying to outsource pairing operations in a secure manner. But besides pairing operations, there are other basic and costly operations in pairing-based cryptography and identity-based schemes, including scalar multiplication on elliptic curves. In this research, we outsource the Boneh-Franklin encryption in a more secure and efficient (in terms of computational and communication complexity) way than existing schemes. Also we outsource the BLMQ signature (based on Sakai-Kasahara) scheme for the first time. The proposed schemes are secure in the OMTUP model. Also, unlike previous schemes, we considered communication channels insecure. Moreover, compared with the trivial solution which outsources every single operation (such as pairing, scalar multiplication and modular exponentiation) as a separate subroutine, our schemes offer less complexity by seamlessly outsourcing the whole encryption scheme for the first time.
Research Article
Farhad Taheri Ardakani; Siavash Bayat Sarmadi
Abstract
Secure multi-party computation (MPC) allows a group of parties to compute a function on their private inputs securely. Classic MPC protocols for two parties use either Yao's garbled circuit (GC) or the Goldreich-Micali-Wigderson (GMW) protocol. In this paper, we propose MISC, a multi-input secure computation ...
Read More
Secure multi-party computation (MPC) allows a group of parties to compute a function on their private inputs securely. Classic MPC protocols for two parties use either Yao's garbled circuit (GC) or the Goldreich-Micali-Wigderson (GMW) protocol. In this paper, we propose MISC, a multi-input secure computation protocol, by combining GC and GMW in a novel way. MISC can evaluate multi-input AND gates, which can reduce the round complexity. Moreover, MISC reduces the communication overhead by 1.7x and 2.4x for 2-input and by 2x and 2.8x for 4-input AND gates compared to the state-of-the-art GMW-style and GC-style protocols, respectively. In order to use the MISC efficiently in different applications, we redesign common building block with multi-input AND gates such as Equality checking, Maxpool, Comparison, and Argmax/Argmin. Results on privacy-preserving applications, e.g., circuit-based private set intersection (PSI) and private machine learning (CNN inference) show that compared to GMW, MISC improves the total communication overhead by 3x and the total run time by 1.5x.