Grad student wins best student paper at international signal processing conference
May 19, 2026
UNIVERSITY PARK, Pa. — Guangmingmei Yang, a computer science and engineering doctoral student at Penn State, has received the Best Student Paper Award at the 2026 IEEE International Conference on Acoustics, Speech, and Signal Processing (ICASSP) for research advancing machine learning security and robustness.
The award recognizes Yang’s paper, “Improving maximum margin backdoor detection by class subspace decorrelation,” which addresses vulnerabilities in machine learning models caused by backdoor attacks. Yang’s work focuses on improving detection methods by reducing correlations between class subspaces, strengthening the reliability of model behavior under adversarial conditions.
Yang is advised by David J. Miller, professor of electrical engineering, and George Kesidis, professor of computer science and engineering, both of whom contributed to the collaborative research effort.
Share this story:
