Efficient multi-key homomorphic evaluation for applications in genomics

• Grant number: 21/09849-5
• Support Opportunities: Scholarships abroad - Research Internship - Doctorate
• Start date: February 01, 2022
• End date: January 31, 2023
• Field of knowledge: Physical Sciences and Mathematics - Computer Science - Theory of Computation
• Principal Investigator: Edson Borin
• Grantee: Antonio Carlos Guimarães Junior
• Supervisor: Diego de Freitas Aranha
• Host Institution: Instituto de Computação (IC). Universidade Estadual de Campinas (UNICAMP). Campinas , SP, Brazil
• Institution abroad: Aarhus University, Denmark
• Associated to the scholarship: 19/12783-6 - Efficient migration of high-performance computing science and engineering applications to the cloud, BP.DR

Abstract

Human genome data are sensitive information that requires strong guarantees of privacy. This issue is aggravated by the increasing use of cloud computing to process it, as the cloud is used not only for computation outsourcing but also to provide services and perform multi-party protocols. For many applications, it is often necessary to combine data from multiple providers, which represents an additional challenge since each provider might have its own privacy restrictions that must be met simultaneously. This problem could be solved using classical Multi-party Computation (MPC), but those are interactive protocols with high communication costs. A recent alternative is the use of Multi-Key Fully Homomorphic Encryption (MKFHE), which enables the possibility of performing computation over data encrypted by different parties. While promising in theory, practical solutions are very recent and still incur significant performance penalties. In this project, we intend to implement, evaluate, and improve the most recent MKFHE techniques for the TFHE cryptosystem (Chillotti et al., JoC 33(1):34-91, 2020). We also intend to implement and compare alternative approaches, such as threshold cryptography and proxy re-encryption to other cryptosystems. We plan on performing such evaluation considering the specific contexts and necessities of our applications of interest, although we also expect more broad (nonspecific to application context) contributions. (AU)