§ feed · storyline
Testing robustness against unforeseen adversaries
Researchers publish a method to evaluate neural network classifier robustness against unseen adversarial attacks, introducing the UAR metric to measure performance across diverse unforeseen attack types.
We’ve developed a method to assess whether a neural network classifier can reliably defend against adversarial attacks not seen during training. Our method yields a new metric, UAR (Unforeseen Attack Robustness), which evaluates the robustness of a single model against an unanticipated attack, and highlights the need to measure performance across a more diverse range of unforeseen attacks.
§ sources1 publication · timeline below