Bridging Strong Privacy with Scalability for Online Data Services

Abstract: Privacy over the Internet has become a huge concern over the past few years. The vast volume of data we disclose to various online services is susceptible to threats from both internal and external attackers. Moreover, powerful entities such as nation-states or government agencies collect user data for mass surveillance. In this talk, I will discuss how we can redesign and rebuild various widely used online services to protect our data against all these types of attacks. Throughout the discussion, I will highlight the enormous tension between strong privacy and scalability prevailing in state-of-the-art solutions. Next, I will demonstrate how we can tailor our system designs, considering the strengths and weaknesses of different privacy-preserving building blocks, to ensure scalability while offering robust privacy guarantees to users. Finally, I will summarize my future research vision, which involves building scalable privacy preserving solutions in the areas of databases and federated learning.

Bio: Ishtiyaque Ahmad is a PhD candidate in the Department of Computer Science at University of California, Santa Barbara. His research interests lie at the intersection of distributed systems, data management, and cryptography. More specifically, his research focuses on building large-scale data systems that prioritize strong privacy for end-users. The outcomes of his research have been published at some of the most reputed venues, including ACM SOSP, USENIX OSDI, VLDB, and ISCA. He is a recipient of an IBM PhD Fellowship and an Outstanding Paper award from the UCSB Computer Science Department.

A pizza lunch for attendees will be available at 11:45 a.m. in Computer Science Building, Room 150/151.