OpenAI 提出 UAR 指标评估模型对抗未知攻击鲁棒性
原标题:Testing robustness against unforeseen adversaries
45AI 研判分
研究人员提出“未知攻击鲁棒性”(UAR)新指标,用于评估神经网络分类器在面对训练时未见过的对抗攻击时的防御可靠性。该方法强调需通过更多样化的未知攻击来衡量模型性能,揭示了现有单一基准测试的局限性。
报道正文
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.