FRG Biweekly Meeting: Yao Li, UNC

Title:Security of Deep Learning Models

Speaker: Yao Li, UNC

Abstract: In this talk, Dr. Li introduced two of her recent works in the field of security of deep learning models. The first one is “Detecting Adversarial Examples with Bayesian Neural Network”, which is about testing time security. The second one is called “Trusted Aggregation (TAG): Model Filtering Backdoor Defense in Federated Learning”, which is about training time security.