Distributional Robustness Loss for Long-tail Learning
Open paper
Benchmark
Model
Rank
Results
long-tail-learning-on-cifar-100-lt-r-10
DRO-LT
–
Error Rate: 36.59
long-tail-learning-on-cifar-100-lt-r-100
DRO-LT
–
Error Rate: 52.67
long-tail-learning-on-imagenet-lt
DRO-LT
–
Top-1 Accuracy: 53.5
Rank counts only results with a code link.