EE Seminar: Addressing the privacy and energy efficiency challenges of largescale information systems
Thu, Jun 14, 2018 @ 10:30 AM - 11:30 AM
Ming Hsieh Department of Electrical and Computer Engineering
Conferences, Lectures, & Seminars
Speaker: Dr. Peter Kairouz, Postdoctoral Research Fellow/Stanford University
Talk Title: Addressing the privacy and energy efficiency challenges of largescale information systems
Abstract: The explosive growth in connectivity and information sharing across a multitude of sensory devices has been accelerating the use of machine learning to guide consumers through a myriad of choices and decisions. While this vision is expected to generate many disruptive business and social opportunities, it presents a number of unprecedented challenges. My talk will address two of these challenges: sharing largescale datasets in a privacy-preserving fashion, and enabling a massive number of sporadically active low-energy wireless devices with small payloads to access the spectrum with minimal coordination and channel estimation overheads.
In the first part of my talk, I will present fundamental (and somewhat surprising) results on sparse group testing, a version of the classical group testing problem with a constraint on the number of tests an item is allowed to participate in. I will also show how these results aid in the design of low-energy random access protocols.
In the second part of my talk, I will introduce a novel privacy notion called generative adversarial privacy (GAP). GAP leverages recent advancements in adversarial learning to arrive to a unified framework for data-driven privacy that has deep game-theoretic and information-theoretic roots. I will also showcase the performance of GAP on real-life datasets.
I will conclude my talk by discussing exciting future research directions.
Biography: Peter Kairouz is a postdoctoral research fellow at Stanford University. He received his Ph.D. in ECE, M.S. in Maths, and M.S. in ECE from the University of Illinois at Urbana-Champaign (UIUC) and his B.E. in ECE from the American University of Beirut (AUB). He interned twice at Qualcomm and more recently at Google where he designed privacy-aware unsupervised learning algorithms. He is the recipient of the 2012 Roberto Padovani Scholarship from Qualcomm's Research Center, the 2015 ACM SIGMETRICS Best Paper Award, and the 2016 Harold L. Olesen Award for Excellence in Undergraduate Teaching from UIUC. His research interests are interdisciplinary and span the areas of data and network sciences, privacy-preserving data analysis, machine learning, and information theory.
Host: Dr. Keith Chugg, email@example.com
Audiences: Everyone Is Invited
Contact: Mayumi Thrasher