Table 4 Spearman's rank-order correlation coefficient of ranking results based on MNIST-C and CIFAR-10-C. For Random, SD, and CES, we compute the average and standard deviation over all labeling budgets and 50-time experiments. The best performance is highlighted in gray. The values highlighted in yellow are where CES or random outperform SDS. The higher the better.

Table 5 Kendall's tau of ranking results based on MNIST-C and CIFAR-10-C. 

Table 6 Jaccard similarity of ranking the top-k DNNs based on the robustness. For Random, SDS, and CES, we report the average results over all labeling budgets. The best performance is highlighted in gray. The higher the better.