Allon Adir, Ehud Aharoni, Nir Drucker, Ronen Levy, Hayim Shaul, Omri Soceanu

Homomorphic Encryption for Data Science (HE4DS) (eBook, PDF)

• Format: PDF

Produktbeschreibung

This book provides basic knowledge required by an application developer to understand and use the Fully Homomorphic Encryption (FHE) technology for privacy preserving Data-Science applications. The authors present various techniques to leverage the unique features of FHE and to overcome its characteristic limitations.

Specifically, this book summarizes polynomial approximation techniques used by FHE applications and various data packing schemes based on a data structure called tile tensors, and demonstrates how to use the studied techniques in several specific privacy preserving applications. Examples and exercises are also included throughout this book.

The proliferation of practical FHE technology has triggered a wide interest in the field and a common wish to experience and understand it. This book aims to simplify the FHE world for those who are interested in privacy preserving data science tasks, and for an audience that does not necessarily have a deep cryptographic background, including undergraduate and graduate-level students in computer science, and data scientists who plan to work on private data and models.

---

Dieser Download kann aus rechtlichen Gründen nur mit Rechnungsadresse in A, B, BG, CY, CZ, D, DK, EW, E, FIN, F, GR, HR, H, IRL, I, LT, L, LR, M, NL, PL, P, R, S, SLO, SK ausgeliefert werden.

Produktdetails

• Verlag: Springer International Publishing
• Seitenzahl: 304
• Erscheinungstermin: 9. November 2024
• Englisch
• ISBN-13: 9783031654947
• Artikelnr.: 72350718

Autorenporträt

Allon Adir holds an M.Sc. in Computer Science from the Technion - Israel Institute of Technology, and is now a researcher in the AI Security group at IBM Research - Israel in Haifa. Allon worked on research related to hardware verification, and then on applications of analytics to Cyber-security. Allon is currently working on novel encryption schemes and their application in the context of security and privacy preservation. Allon has authored many publications and patents related to the above fields, and is an IBM Master Inventor.

Ehud is currently with IBM Research in Haifa, Israel. Ehud received an M.Sc. in computer science from the Technion - Israel Institute of Technology. He worked several years on machine learning projects in the fields of hardware verification, healthcare, and anomaly detection for computer systems. Later, he worked on various applications of machine learning to cyber security. Ehud is currently working on novel encryption schemes in the context of security and privacy preservation.

Nir Drucker is a Security, Privacy & Cryptography Research Scientist at IBM Research - Israel, the AI Security group that develop the IBM HElayers SDK. He holds a Ph.D. in Applied Mathematics (Cryptography) from the University of Haifa and an M.Sc. degree in Operations Research from the faculty of Industrial Engineering & Management of the Technion I.I.T. Nir worked 3.5 years as a Senior Applied Scientist in AWS, the Cryptographic Algorithms Team, and eight years as a Software Developer at Intel in two teams: a) a team that developed low-level security features (e.g., SGX) in FW/SW; b) a team that developed a CAD VLSI timing verification simulator in C/C++. His research interests involve applied cryptography and applied security. In particular, research that combines these domains with the latest development in the machine-learning field. For example, researching Privacy-Preserving Machine Learning (PPML) solutions that involve Homomorphic Encryption (HE) or multi party computation (MPC).

Ronen Levy is a senior manager, leading the Security & Privacy department in the IBM Research - Israel lab also responsible for the IBM Research strategy around Data Security. He holds a B.A in Mathematics from the University of Haifa and with over 30 years of experience in R&D he has been driving innovation in various domains such as hybrid-cloud, software quality, privacy, cyber-security and cryptography. Before joining IBM Research 14 years ago he worked in various R&D development roles in the industry developing products such as Anti-Virus, Distributed Query Processor, Application Server and Enterprise Storage System.

Hayim is currently with IBM Research in Haifa, Israel. Hayim completed his PhD in computational geometry under the supervision of Prof. Micha Sharir in Tel- Aviv University. Following his studies, Hayim cofounded and served as CTO of DiviNetworks, a company funded by the IFC, that focused on network optimizations. After that, he joined MIT as a research fellow in the CSAIL lab doing research in homomorphic encryption. At IBM Research Hayim is continuing his work on secure multi-party computation.

Omri Soceanu is the head of the AI Security research group at IBM Research Haifa. He holds a B.Sc. and M.Sc. in Electrical Engineering from the Technion - Israel Institute of Technology. Before becoming the head of the AI Security group, Omri worked on various aspects of data security, employing machine learning techniques in a Big Data setting using state-of-the-art approaches. Omri has several years of hands-on experience working on embedded systems, cryptography, cybersecurity, and machine learning algorithms.

Inhaltsangabe

Part I Introduction and Basic Homomorphic Encryption (HE) Concepts.- Chapter 1 Introduction to Data Science.- Chapter 2 Modern Homomorphic Encryption - Introduction.- Chapter 3 Modern HE - Security Models.- Chapter 4 Approaches for Writing HE Applications.- Part II Approximations.- Chapter 5 Approximation Methods Part I: A General Overview.- Chapter 6 Approximation Methods Part II: Approximations of Standard Functions.- Part III Packing Methods.- Chapter 7 SIMD Packing Part I: Basic Packing Techniques.- Chapter 8 SIMD Packing Part II - Tile Tensor Basics.- Chapter 9 SIMD Packing Part III: Advanced Tile Tensors.- Part IV Use Cases and Other Approaches.- Chapter 10 Privacy-Preserving Machine Learning with HE.- Chapter 11 Case Study: Neural Network.