Long-Tailed Recognition via Weight Balancing

Open paper
Benchmark Model Rank Results
long-tail-learning-on-cifar-100-lt-r-10LTR-weight-balancing#9Error Rate: 31.33
long-tail-learning-on-cifar-100-lt-r-100LTR-weight-balancing#15Error Rate: 46.45
long-tail-learning-on-cifar-100-lt-r-50LTR-weight-balancing#13Error Rate: 42.29
long-tail-learning-on-imagenet-ltLTR-weight-balancing(ResNeXt-50)#40Top-1 Accuracy: 53.9
long-tail-learning-on-inaturalist-2018LTR-weight-balancing(ResNet-50)#31Top-1 Accuracy: 70.2%

Rank counts only results with a code link.