Teaching AI What They Don't Know: Foundations and Algorithms for Reliable ML in the Open World

Speaker: Sean (Xuefeng) Du
University of Wisconsin-Madison

Title: "Teaching AI What They Don't Know: Foundations and Algorithms for
Reliable ML in the Open World"

Date: Thursday, 19 December 2024

Time: 10:00am via Zoom

Join Zoom Meeting
https://hkust.zoom.us/j/96688516988?pwd=qfj1PQIjEi0I75lwVGfY7PurdPDRBW.1

Meeting ID: 966 8851 6988
Passcode: 202526

Abstract:

The remarkable capabilities of machine learning (ML) models, especially 
foundation models like GPT, have transformed numerous domains. However, these 
systems often falter in real-world settings, where they encounter unknown or 
out-of-distribution (OOD) inputs, and generate overconfident predictions or 
unreliable outputs. Ensuring their reliability is not only a technical 
challenge but also a fundamental requirement for their safe deployment.

In this talk, I will discuss my research on teaching ML models what they 
don’t know by developing foundational frameworks for reliable decision-making 
in the open world. This involves three core aspects: (1) designing novel 
algorithms for unknown-aware learning through adaptive outlier synthesis, 
enabling models to handle unfamiliar inputs without explicit knowledge of 
unknowns; (2) leveraging unlabeled data in the wild to detect and generalize 
across diverse real-world reliability challenges; and (3) addressing 
reliability blind spots in foundation models, such as hallucinations, 
malicious prompts, and noisy alignment data, through innovative mitigation 
strategies

Through fundamental algorithmic development, theoretical insights, and 
practical applications, my research contributes to the responsible deployment 
of AI technologies. The talk will conclude with a forward-looking perspective 
on interdisciplinary collaborations and the roadmap for achieving robust, 
reliable AI systems that adapt to an ever-changing world.


***************
Biography:

Sean (Xuefeng) Du is a final year CS Ph.D. student at University of 
Wisconsin-Madison. His research interest is in reliable machine learning and 
the applications to foundation models and AI safety. His first-author papers 
have been recognized with multiple oral and spotlight presentations at 
NeurIPS and CVPR. He is a recipient of the Jane Street Graduate Research 
Fellowship, and Rising Stars in Data Science award.