Paper
in
Workshop: Security and Safety in Machine Learning Systems
Measuring Adversarial Robustness using a Voronoi-Epsilon Adversary
Hyeongji Kim
Abstract: