Learning With Noisy Labels On Cifar 10N 2

Accuracy (mean)

评测结果

各个模型在此基准测试上的表现结果

比较表格

模型名称	Accuracy (mean)
generative-noisy-label-learning-by-implicit	91.42
co-teaching-robust-training-of-deep-neural	90.30
peer-loss-functions-learning-from-noisy	88.76
generalized-cross-entropy-loss-for-training	87.70
learning-with-instance-dependent-label-noise-1	94.88
early-learning-regularization-prevents	94.20
making-deep-neural-networks-robust-to-label	86.28
190600189	87.71
combating-noisy-labels-by-agreement-a-joint	90.21
robust-training-under-label-noise-by-over	95.31
learning-with-instance-dependent-label-noise-1	89.91
making-deep-neural-networks-robust-to-label	86.14
early-learning-regularization-prevents	91.61
psscl-a-progressive-sample-selection	96.21
dividemix-learning-with-noisy-labels-as-semi-1	90.90
模型 16	86.46
clusterability-as-an-alternative-to-anchor	90.75
does-label-smoothing-mitigate-label-noise	89.35
understanding-generalized-label-smoothing	90.37
imprecise-label-learning-a-unified-framework	95.04
when-optimizing-f-divergence-is-robust-with-1	89.79
provably-end-to-end-label-noise-learning	88.27
how-does-disagreement-help-generalization	89.47