Y
Yingyuan Yang
Researcher at University of Illinois at Springfield
Publications - 14
Citations - 110
Yingyuan Yang is an academic researcher from University of Illinois at Springfield. The author has contributed to research in topics: Authentication & Password. The author has an hindex of 5, co-authored 13 publications receiving 61 citations. Previous affiliations of Yingyuan Yang include University of Tennessee.
Papers
More filters
Posted Content
ConAML: Constrained Adversarial Machine Learning for Cyber-Physical Systems
TL;DR: Constrained Adversarial Machine Learning (ConAML), which generates adversarial examples that satisfy the intrinsic constraints of the physical systems, is proposed, which significantly decrease the performance of the ML models even under practical constraints.
Journal ArticleDOI
PersonaIA: A Lightweight Implicit Authentication System Based on Customized User Behavior Selection
TL;DR: This work proposes a W-layer, an overlay that provides a practical and energy-efficient solution for implicit authentication on mobile devices and implemented partially labeled Dirichlet allocation (PLDA) on the server side for more accurate feature extraction.
Proceedings ArticleDOI
Energy-efficient W-layer for behavior-based implicit authentication on mobile devices
Yingyuan Yang,Jinyuan Sun +1 more
TL;DR: This work proposes a W-layer, an overlay that provides an energy-efficient solution for real-time implicit authentication on mobile devices and conducts several experiments on both synthetic and real datasets to evaluate the method.
Proceedings ArticleDOI
SearchFromFree: Adversarial Measurements for Machine Learning-based Energy Theft Detection
TL;DR: In this paper, an adversarial measurement generation approach that enables the attacker to report extremely low power consumption measurements to utilities while bypassing the ML energy theft detection was proposed. But the attack was limited to black-box attacks.
Proceedings ArticleDOI
Retraining and Dynamic Privilege for Implicit Authentication Systems
TL;DR: The proposed techniques can successfully detect the degradation of accuracy of the user behavior model, as well as automatically determine and adjust to the best retraining frequency, and reduces the impact of false negatives on legitimate users and enhances system reliability and user experience.