CRC Seminar Series - Ilaria Chillotti

Title:

Introduction to FHE, TFHE and Machine Learning applications

Abstract:

FHE is a technology allowing to perform operations over encrypted data. In this talk, we will do a general introduction to the existing FHE schemes, followed by a more detailed description of the TFHE scheme. By enabling both leveled and bootstrapped operations, TFHE can be used for a wide range of use-cases, from homomorphic Boolean circuits to homomorphic neural networks. In the final part of the talk, we will concentrate on how to use TFHE to evaluate neural network inferences over encrypted data.