Secure multiparty computation (MPC), fully homomorphic encryption (FHE), federated learning (FL), trusted execution environments (TEEs) and differential privacy (DP) are prominent examples of emerging PETs. They enable the computation of a function without revealing the input data. By incorporating these techniques, organizations strike a balance between preserving the privacy of sensitive input data and deriving valuable insights from data analysis, optimizing the privacy-utility tradeoff.
This tutorial will walk you through these technologies and show how to combine them in order to implement a hybrid PETs use-case. In the practical session you will learn how to implement a complete and hybrid privacy-preserving machine learning workflow using different PETs tools and libraries for PSI, MPC, FHE amongst others.
|Lecture 1 (40’)||Introduction PETs, Computing on Encrypted Data (TBC): Nigel Smart (KU Leuven, Zama)|
|Lecture 2 (40’)||Introduction MPC, FL: (Inpher)|
|Lecture 3 (40’)||Introduction FHE: Nicolas Gama (SandboxAQ)|
|2h40 afternoon||Practical sessions on hybrid PETs ML (MPC, FHE, FL) workflows using different tools and libraries (from different industrial and academical bodies)|
|17h00-20h||Dinner / Cocktails|
Mariya Georgieva leads Inpher’s cryptography research and manages a team of engineers. She is responsible for developing "Secret Computing'' technology and for proposing, designing, and developing privacy preserving solutions, including MPC, FHE, and FL. Dr. Georgieva is co-author of the open-source TFHE library.
Sergiu Carpov is a senior cryptographer at Inpher and his main research interests include efficient secure computation protocols and compilation techniques for privacy-preserving applications.
132 West 31st Street, 9th Floor, New York, NY 10001