2/22/2023 Aaron Seidlitz, Illinois CS
As cryptography continues to evolve, Illinois CS professor Dakshita Khurana positions her research to improve its impact on security while maintaining privacy and efficiency.
Written by Aaron Seidlitz, Illinois CS
A Catch-22 stands in the way of computer scientists truly understanding what it means for a system to be secure.
As Illinois Computer Science professor Dakshita Khurana describes it, proving that a proposed system resists attacks from arbitrarily malicious adversaries means the operator must show how to extract implicit inputs from the adversary and inspect them for consistency.
That process of input extraction, however, is at odds with other goals like privacy and efficiency.
The ability to reconcile these contradictory objectives has led researchers to further delve into the expanding purpose of cryptography. And that just so happens to be one of Khurana’s primary research interests.
Her recent proposal on the topic, “Cryptographic Proofs, Outside the Black-Box,” proved promising and earned her the NSF CAREER Award earlier this year.
“Within this beautiful landscape of cryptography that we have today, there is still a lot we do not understand. I hope to change some of that going forward,” Khurana said. “The methods developed as part of this work should have broad applications in the design of simpler, secure, and trustless cryptosystems.”
This increased role cryptography plays in security hasn’t always been the case.
In fact, Khurana points this out in her NSF CAREER Award proposal. As it originated, she states, cryptography focused first on securing communications. Projects like hers are expanding it well beyond its original goal.
Today, it allows researchers and practitioners an outlet to compute on data in provably secure ways. And cryptographic proofs, the point of emphasis in her CAREER Award proposal, “form the bedrock of security” against adversaries and have facilitated powerful capabilities.
A primary breakthrough in this effort dates back 30 years, to a discovery made by Shafi Goldwasser, Silvio Micali, and Charles Rackoff.
“By allowing a verification algorithm to interactively send random challenges to a prover, it became possible to prove statements while also hiding secrets about them,” Khurana said of the that discovery. “And yet, despite three decades of research and wide adoption in theory and practice, there are some basic questions about cryptographic proofs that remain mysteries to this day. For example, we still do not understand whether these proofs really require interaction, and my CAREER proposal aims to remedy this situation.
“Assuming that certain well-studied mathematical problems are hard to solve, we will aim to remove interaction while continuing to hide secrets in proofs. We will also aim to significantly expand the class of computations that admit efficiently verifiable proofs. This may require new, out-of-the-box reasoning techniques which carefully inspect the code of an adversary that attempts to attack a system.”
In addition, Khurana also plans to build a research community to aid in this effort. Workshops and training sessions will include graduate, undergraduate, and K-12 students.
She appreciates the way this project continues to offer chances at inspiration alongside her students. Simultaneously, it will showcase the way in which mentorship within Illinois CS leads to success.
“While these questions have captivated me for many years, graduate students at the University of Illinois Urbana-Champaign have also recently engaged in some fantastic research on zero-knowledge and succinct proofs,” Khurana said. “The CAREER award will help me support these students as they continue to advance in the field. I also hope to involve undergraduate students in this research via course projects and internship opportunities. These funds will also help us set up a series of cryptography workshops to bring together researchers in the Midwest.
“I am thankful to the department, especially Department Chair Nancy Amato, and fellow CS professors Tandy Warnow, Chandra Chekuri, Ruta Mehta and Sariel Har-Peled for their constant support and guidance. I am also grateful to be surrounded by many kind colleagues in the cryptography community who took the time to read my drafts and provide valuable feedback.”