Combating noisy labels by agreement: A joint training method with co-regularization

Benchmark Model Rank Results
learning-with-noisy-labels-on-cifar-10n-random3JoCoR#11Accuracy (mean): 90.11
learning-with-noisy-labels-on-cifar-100nJoCoR#12Accuracy (mean): 59.97
learning-with-noisy-labels-on-cifar-10n-random1JoCoR#12Accuracy (mean): 90.30
learning-with-noisy-labels-on-cifar-10n-random2JoCoR#12Accuracy (mean): 90.21
learning-with-noisy-labels-on-cifar-10n-worstJoCoR#14Accuracy (mean): 83.37
learning-with-noisy-labels-on-cifar-10n-aggregateJoCoR#15Accuracy (mean): 91.44
image-classification-on-clothing1mJoCoR#43Accuracy: 70.3%