Close

Session

Research Manuscript: Unleashing Privacy for AI
Event TypeResearch Manuscript
Topics
AI
Keywords
AI/ML Security/Privacy
TimeTuesday, July 11th1:30pm - 3:00pm PDT
Location3004, 3rd Floor
DescriptionNavigating the challenges of AI and tremendous data and protecting privacy and intellectual property are the topics of this session. It focuses on the complex interplay between artificial intelligence (AI), large and complex data sets, and privacy. The session will present several interesting and exciting works that aim at building privacy-enhancing AI. Whether you are a business leader, data scientist, or privacy advocate, this session will provide valuable insights and practical solutions for ensuring that AI and data are used responsibly and in compliance with privacy.
The session will delve into the latest advancements in privacy-enhancing methods and deep learning. The first paper, PASNet, presents a framework for secure two-party deep learning with low latency and high accuracy. The second paper, STML, focuses on secure IP for Tiny ML microcontrollers with optimized security and latency. The third paper proposes a new approach for privacy enhancement in machine learning using empirically-defined privacy evaluation. Finally, the fourth paper introduces Primer, a fast and accurate Transformer for NLP using encrypted data.