Command Palette
Search for a command to run...
Detection of Area Under the ROC Curve Det AUC
Date
Tags
Detection Area Under the ROC Curve is an important evaluation metric for adversarial example detection tasks in the field of AI safety and robustness. It measures the overall discriminative ability of a detector across all thresholds by calculating the area under the ROC curve when the detector distinguishes adversarial examples from normal examples. This metric is insensitive to the choice of a single threshold, facilitating fair comparisons among different detection methods. It is usually reported as a dimensionless score in the interval [0, 1], where a higher value indicates stronger detection and discrimination ability, and it is widely adopted in relevant benchmarks and papers.
Build AI with AI
From idea to launch — accelerate your AI development with free AI co-coding, out-of-the-box environment and best price of GPUs.