Explore a 19-minute IEEE conference talk on enhancing neural network security through model orthogonalization and class distance hardening. Delve into the research conducted by a team from Purdue University, including Guanhong Tao, Yingqi Liu, Guangyu Shen, Qiuling Xu, Shengwei An, Zhuo Zhang, and Xiangyu Zhang. Learn about innovative techniques to improve the robustness and reliability of neural networks in the face of security challenges, and gain insights into the potential applications of these methods in various fields of artificial intelligence and machine learning.
Model Orthogonalization - Class Distance Hardening in Neural Networks for Better Security
Overview
Syllabus
Model Orthogonalization: Class Distance Hardening in Neural Networks for Better Security
Taught by
IEEE Symposium on Security and Privacy
Tags
Related Courses
-
Model Stealing Attacks Against Inductive Graph Neural Networks
-
SoK - Certified Robustness for Deep Neural Networks
-
Distillation as a Defense to Adversarial Perturbations against Deep Neural Networks
-
Towards Evaluating the Robustness of Neural Networks
-
WtaGraph - Web Tracking and Advertising Detection using Graph Neural Networks
-
The Cost of Learning from the Best - How Prior Knowledge Weakens the Security of Deep Neural Networks
