Computer Engineering Seminar Series

We will examine (1) new frameworks for evaluating and aligning model behavior with human intent (2) the security and reliability of watermarking techniques in foundation models, including their role in provenance tracking and their vulnerabilities to adversarial removal and evasion, and (3) novel approaches for detecting and mitigating high-risk model outputs before deployment. By synthesizing these findings, we will discuss the broader implications of foundation model security, trade-offs between robustness and control, and future directions for improving Al safety at scale.

The focus of the talk, however, will be a measurement study we performed of the University of Chicago’s vulnerability to credential-guessing attacks across twenty years. Given a list of university usernames, we searched for matches in data breaches from hundreds of websites. Ultimately, we successfully guessed passwords for thousands of UChicago affiliates. I will conclude my talk by discussing our parallel investigations of FIDO2 passwordless authentication, including analyzing why passkeys have not yet replaced passwords for web authentication.

In the first part of this talk, I will discuss our efforts to apply hardware-software co-design to help future heterogeneous systems overcome these challenges and improve performance, energy efficiency, and scalability. Then, in the second part I will discuss how the on-going transition to chiplet-based heterogeneous systems exacerbates these challenges and how we address these challenges in chiplet-based heterogeneous systems by rethinking the control plane.

As ML on structured data becomes prevalent across enterprises, improving resource efficiency is crucial to lower costs and energy consumption. Designing systems for learning on structured data is challenging because of the large number of models. Parameters and data access patterns. We identify that current systems are bottlenecked by data movement which results in poor resource utilization and inefficient training.

In this talk, I will first discuss our recent work on characterizing the inherent tradeoff between fairness and accuracy in both classification and regression problems, where we show that the cost of fairness could be characterized by the optimal value of a Wasserstein-barycenter problem. Then I will show that the complexity of learning the optimal fair predictor is the same as learning the Bayes predictor and present a post-processing algorithm based on the solution to the Wasserstein-barycenter problem that derives the optimal fair predictors from Bayes score functions

Modern cloud-based applications adhere to a microservices architecture, encompassing numerous components interconnected through intricate dependencies, operating within a distributed environment. This talk gives an overview of three recent research projects aimed at solving practical challenges in promptly identifying and diagnosing outages.