Towards Unlocking the Mystery of Adversarial Fragility of Neural Networks

Friday, February 16, 2024 - 3:00pm

Event Calendar Category

LIDS Seminar Series

Speaker Name

Weiyu Xu

Affiliation

University of Iowa

Building and Room Number

32-D677

Abstract

In supervised learning, neural network based classifiers have been observed to be fragile under adversarial attacks. In this talk, we present an information-theoretic and then a matrix-theoretic explanation for this phenomenon. 

Biography

Weiyu Xu received his B.E. in Information Engineering from Beijing University of Posts and Telecommunications in 2002, and a M.S. degree in Electronic Engineering from Tsinghua University in 2005. He received a M.S. and a Ph.D. degree in Electrical Engineering in 2006 and 2009 from California Institute of Technology (Caltech), with a minor in Applied and Computational Mathematics.

He is currently an associate professor at the Department of Electrical and Computer Engineering at the University of Iowa. His research interests are signal processing, optimization, and high dimensional data analytics.

https://engineering.uiowa.edu/people/weiyu-xu