Universal Adversarial Perturbations in Machine Learning - Spring 2021

Explore universal adversarial perturbations in this 23-minute lecture from the University of Central Florida's CAP6412 course. Delve into the motivation behind these perturbations, their definition, and key contributions to the field. Examine experimental results, visualizations, and cross-model universality. Gain insights into the process of finetuning and its implications for adversarial machine learning.

Introduction
Outline
Motivation
Definition
Contributions
Universal perturbations
Experiments
Results
Visualization
Crossmodel universality
Visualizations
Finetuning