Concurrent Encryption and Authentication

Prof. Josef Pieprzyk  |  February 28, 2020

Signcryption aims to provide both confidentiality and authentication of messages more efficiently than the independent application of encryption and signature. In the talk, we consider a generic and efficient signcryption scheme featuring parallel encryption and signature on top of a sponge-based message-padding underlying structure. Unlike other existing schemes, the proposed scheme also supports arbitrary long messages. We prove the construction secure when instantiated from weakly secure asymmetric primitives such as a trapdoor one-way encryption and a universal unforgeable signature.

Secure Multiparty Computation versus Secure Outsourcing

Prof. Yvo Desmedt  |  February 25, 2020

The talk starts with explaining Secure Multi-Party Computation (MPC). Then we explain the rise of cloud storage, cloud computing and social networks. We regard it as a consequence of a failure in the design of adequate OS (operating systems). We survey some of the solutions proposed to address security problems on the cloud. This presentation is focused on analyzing whether MPC is the correct technique for this problem. Moreover, besides issues as speed, we show that other problems pop up that are irrelevant in a typical MPC setting.

Synthetizing Privacy-Preserving Time Use Traces

Prof. Eran Toch  |  November 20, 2019

Synthetizing data that reflects people's behavior traces is one of the most compelling methods for large-scale privacy-preserving data analysis. However, the task is challenging because of the complexity in generating accurate and useful behavioral traces, in complex domains such as mobility patterns or electronic health records. In this talk, I will present our work on synthesizing time use data that (derived from mobile phone logs.) We suggest several criteria for assessing the quality of the generated data, such as diversity, similarity to the original data, and statistical resemblance. We then compare Generative Adversarial Networks with other methods and discuss the implications of trace generation to solve troubling privacy problems.

Efficient Information-Theoretic Secure Multiparty Computation Over Z/p^k Z via Galois Rings

Mark Abspoel  |  November 20, 2019

At CRYPTO 2018, Cramer et al. introduced a secret-sharing based protocol called SPDZ2k that allows for secure multiparty computation (MPC) in the dishonest majority setting over the ring of integers modulo 2^k, thus solving a long-standing open question in MPC about secure computation over rings in this setting. In this work we study this problem in the information-theoretic scenario. More specifically, we ask the following question: Can we obtain information-theoretic MPC protocols that work over rings with comparable efficiency to corresponding protocols over fields? We answer this question in the affirmative by presenting an efficient protocol for robust Secure Multiparty Computation over Z/p^k Z (for *any* prime p and positive integer k) that is perfectly secure against active adversaries corrupting a fraction of at most 1/3 players, and a robust protocol that is statistically secure against an active adversary corrupting a fraction of at most 1/2 players.

MPC Engines and their Applications in Secure & Private Business Transactions

Prof. Ronald Cramer  |  October 25, 2019

In this talk, Professor Cramer will present an introduction to the vibrant area of Secure Multiparty Computation (MPC). This area deals with multiparty processing on mutually private data with the purpose of enabling controlled release of information, in the face of mutual mistrust or conflicting interests and in the absence of a “trusted incorruptible party”. In fact, MPC asks for a network of servers to *emulate* such a trusted party even when such a party is not available and even in the face of a malicious adversary. Its invention and subsequent further development is one of the crowning achievements in modern cryptography. After several decades of research, its theory encompasses a wide range of dedicated cryptographic techniques and results now ready for tackling a host of pressing practical problems of real-life size. Application domains are myriad and include for instance, auctions, voting, machine learning, and benchmarking, all with guaranteed correctness and with maximal protection of private data.

Private Set Intersection

Prof. Benny Pinkas  |  August 29, 2019

Private set intersection (PSI) allows two parties to compute the intersection of their sets without revealing any information about items that are not in the intersection. PSI is relevant in many scenarios of secure computation, such as data sharing or contact discovery. PSI is one of the best-studied applications of secure computation and many different PSI protocols have been proposed, using a wide and interesting variety of cryptographic tools. However, existing PSI protocols do not scale up well, and therefore some applications use insecure solutions instead. This talk will survey what we believe to be the most interesting PSI protocols, describe new approaches for designing PSI protocols, and present a performance comparison.

Introduction to Secure Multi-Party Computation

Prof. Benny Pinkas  |  August 22, 2019

Secure multi-party computation enables different parties with private inputs to compute joint functions of these inputs while hiding everything but the output of the function. As a simple example, consider two parties with private values that wish to compute which of these values is greater while hiding all other information about the values. In recent years there has great progress in the performance of secure multi-party computation, and considerable interest in using this technology for different applications. The talk will describe the basic concepts of secure multi-party computation, as well as different techniques that are used to improve performance, and applications that benefit from this technology.

Privacy-preserving Techniques with Applications in Biomedical Data and Other Areas

Prof. SM Yiu  |  August 14, 2019

Data privacy becomes a major concern of the general public and governments. On the other hand, it is important for researchers and industries to collaborate by contributing their data. A critical question is how to integrate data from multiple parties for analysis while protecting the privacy and confidentiality of the data. A trivial solution is to "anonymize" the data before sharing. But there is no perfect solution for anonymization. In this talk, we try to tackle the problem from another perspective. That is, we encrypt the data and try to perform computation on encrypted data without decryption, i.e., without looking at the raw data, we try to compute useful information from encrypted data provided by multiple parties. We will provide an overview how this can be done and also show some applications, such as biomedical applications and blockchain, that can leverage these techniques.

Introduction to Fully Homomorphic Encryption

Prof Hyung Tae Lee  |  August 7, 2019

Fully homomorphic encryption (FHE) is an encryption scheme that allows arbitrary computations on encrypted data without decrypting it. Since its concept was firstly introduced by Rivest, Adleman and Dertouzos in 1978, there had been many trials to realize a secure FHE scheme due to its attractive applications, but all were failed for 30 years. In 2009, Gentry finally succeeded in designing a secure FHE scheme by introducing a new design strategy and providing an instantiation from ideal lattices. Following Gentry's blueprint, there have been proposed various FHE schemes to improve efficiency. In this talk, I will present Gentry's technique to construct FHE schemes with examples and introduce basic applications of FHE schemes.