Lecture by Shafi Goldwasser, Director of The Simons Institute for the Theory of Computing,  UC Berkeley.

Title: Trusting the ML process: A Cryptographer's perspective

TIME & PLACE: Tuesday, May 21 at 11 am in 2-449

Abstract: For decades now cryptographic tools and models have enabled the use of technology platforms controlled by worst case computationally bounded adversaries. In this talk I will describe how to use cryptographic modeling and tools to build trust in various phases of the machine learning pipelines. We will touch on privacy in the training stage, verification of answers and properties of machine learning models post and during training, and robustness to data distributions different from the training data distribution.