Developing Privacy-preserving AI Systems: The Lessons learned
Authors:
- 1 July 2020
- conference paper
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 1-4
Abstract
Advances in customers' data privacy laws create pressures and pain points across the entire lifecycle of AI products. Working figures such as data scientists and data engineers need to account for the correct use of privacy-enhancing technologies such as homomorphic encryption, secure multi-party computation, and trusted execution environment when they develop, test and deploy products embedding AI models while providing data protection guarantees. In this work, we share the lessons learned during the development of frameworks to aid data scientists and data engineers to map their optimized workloads onto privacy-enhancing technologies seamlessly and correctly.Keywords
This publication has 23 references indexed in Scilit:
- Oblivious Neural Network Predictions via MiniONN Transformations2017
- Model Inversion Attacks that Exploit Confidence Information and Basic Countermeasures2015
- Controlled-Channel Attacks: Deterministic Side Channels for Untrusted Operating Systems2015
- Two Halves Make a Whole2015
- Efficient Garbling from a Fixed-Key Blockcipher2013
- Improved Security for a Ring-Based Fully Homomorphic Encryption Scheme2013
- Fully homomorphic encryption using ideal lattices2009
- Improved Garbled Circuit: Free XOR Gates and Applications2008
- Privacy preserving auctions and mechanism design1999
- How to generate and exchange secrets1986