Uploaded July 2024 | Updated November 2026, 3 hours ago
Institute for Assured Autonomy & Computer Science Seminar Series
January 10, 2023
“Adversarial Robustness and Forensics for Deep Neural Networks”
Ben Y. Zhao, University of Chicago
Despite their tangible impact on a wide range of real world applications, deep neural networks are known to be vulnerable to numerous attacks, including inference time attacks based on adversarial perturbations, as well as training time attacks such as backdoors. The security community has done extensive work to explore both attacks and defenses, only to produce a seemingly endless cat-and-mouse game. In this talk, Ben Y. Zhao will talk about some of his recent work in ML digital forensics. He will begin by summarizing his goal to seek a broader, more pragmatic view towards adversarial robustness beyond the current static, binary views of attack and defense. Like real-world security systems, he believes given sufficient incentive and resources, attackers will eventually succeed in compromising DNN systems. Just as in traditional security realms, digital forensics tools can serve dual purposes: identifying the sources of compromise so that they can be mitigated while also providing a strong deterrent against future attackers. Zhao will present recent results from two papers in this space (published at the 2022 USENIX Security Symposium and the ACM Conference on Computer and Communications Security) that explore the role of post-attack forensics for improving DNN robustness against both poisoning attacks and adversarial examples. Against poisoning attacks, he shows how to use forensic evidence to identify subsets of training data responsible for the attack. Against adversarial examples, he shows how to survive and recover from server breaches that allow attackers to gain full access to proprietary DNN models.
Ben Y. Zhao is the Neubauer Professor of Computer Science at the University of Chicago. Prior to joining UChicago, he was a professor of computer science at the University of California, Santa Barbara. He completed his PhD at the University of California, Berkeley in 2004 and received his BS from Yale University in 1997. He is an ACM fellow and a recipient of an NSF CAREER award, MIT Technology Review’s TR-35 Award (Young Innovators Under 35), ComputerWorld's Top 40 Technology Innovators Award, an Institute of Electrical and Electronics Engineers International Test Conference Early Career Award, and multiple Google Faculty Awards. His work has been covered by media outlets such as The New York Times, The Boston Globe, The Los Angeles Times, MIT Technology Review, The Wall Street Journal, Forbes, Fortune, CNBC, MSNBC, New Scientist, and Slashdot. Zhao has published extensively in the areas of security and privacy, machine learning, networking, and human-computer interaction. He served as Technical Program Committee (Co-)Chair for the 2016 World Wide Web Conference and the 2018 ACM Internet Measurement Conference. He also serves on the steering committee for the ACM Workshop on Hot Topics in Networks (HotNets) and was general co-chair for HotNets 2020.
Institute for Assured Autonomy & Computer Science Seminar Series
January 10, 2023
“Adversarial Robustness and Forensics for Deep Neural Networks”
Ben Y. Zhao, University of Chicago
Despite their tangible impact on a wide range of real world applications, deep neural networks are known to be vulnerable to numerous attacks, including inference time attacks based on adversarial perturbations, as well as training time attacks such as backdoors. The security community has done extensive work to explore both attacks and defenses, only to produce a seemingly endless cat-and-mouse game. In this talk, Ben Y. Zhao will talk about some of his recent work in ML digital forensics. He will begin by summarizing his goal to seek a broader, more pragmatic view towards adversarial robustness beyond the current static, binary views of attack and defense. Like real-world security systems, he believes given sufficient incentive and resources, attackers will eventually succeed in compromising DNN systems. Just as in traditional security realms, digital forensics tools can serve dual purposes: identifying the sources of compromise so that they can be mitigated while also providing a strong deterrent against future attackers. Zhao will present recent results from two papers in this space (published at the 2022 USENIX Security Symposium and the ACM Conference on Computer and Communications Security) that explore the role of post-attack forensics for improving DNN robustness against both poisoning attacks and adversarial examples. Against poisoning attacks, he shows how to use forensic evidence to identify subsets of training data responsible for the attack. Against adversarial examples, he shows how to survive and recover from server breaches that allow attackers to gain full access to proprietary DNN models.
Ben Y. Zhao is the Neubauer Professor of Computer Science at the University of Chicago. Prior to joining UChicago, he was a professor of computer science at the University of California, Santa Barbara. He completed his PhD at the University of California, Berkeley in 2004 and received his BS from Yale University in 1997. He is an ACM fellow and a recipient of an NSF CAREER award, MIT Technology Review’s TR-35 Award (Young Innovators Under 35), ComputerWorld's Top 40 Technology Innovators Award, an Institute of Electrical and Electronics Engineers International Test Conference Early Career Award, and multiple Google Faculty Awards. His work has been covered by media outlets such as The New York Times, The Boston Globe, The Los Angeles Times, MIT Technology Review, The Wall Street Journal, Forbes, Fortune, CNBC, MSNBC, New Scientist, and Slashdot. Zhao has published extensively in the areas of security and privacy, machine learning, networking, and human-computer interaction. He served as Technical Program Committee (Co-)Chair for the 2016 World Wide Web Conference and the 2018 ACM Internet Measurement Conference. He also serves on the steering committee for the ACM Workshop on Hot Topics in Networks (HotNets) and was general co-chair for HotNets 2020.










