List your company
ResearchFormerly WizWand

Image Classification on CIFAR-10-LT Imbalance Ratio 50 (test)

88.5Top-1 AccuracyBalPoE
74.25277.95181.6585.349Nov 27, 2021Dec 29, 2021Jan 31, 2022Mar 4, 2022Apr 6, 2022May 8, 2022Jun 10, 2022
Updated 8mo ago

Evaluation results

MethodLinks
BalPoEBackbone=ResNet-32, Training Setting=Longer training2022.0688.5
NCLBackbone=ResNet-32, Training Setting=Longer training2022.0687.3
BalPoEBackbone=ResNet-32, Training Setting=Standard2022.0686.2
MISLASBackbone=ResNet-32, Training Setting=Standard2022.0685.7
ACEBackbone=ResNet-32, Training Setting=Standard2022.0684.3
UniMix+BayiasBackbone=ResNet-32, Training Setting=Standard2022.0684.3
TSCBackbone=ResNet-32, Imbalance Ratio=502021.1182.9
LDAM-DRWBackbone=ResNet-32, Training Setting=Standard2022.0681.8
KCL†Backbone=ResNet-32, Imbalance Ratio=502021.1181.7
LDAM-DRWBackbone=ResNet-32, Imbalance Ratio=502021.1180.9
CB-FocalBackbone=ResNet-32, Imbalance Ratio=502021.1179.3
CE-DRWBackbone=ResNet-32, Imbalance Ratio=502021.1178.9
CE-DRSBackbone=ResNet-32, Imbalance Ratio=502021.1178.6
CB-CEBackbone=ResNet-32, Imbalance Ratio=502021.1178.1
CEBackbone=ResNet-32, Training Setting=Standard2022.0677.3
LDAMBackbone=ResNet-32, Imbalance Ratio=502021.1176.8
FocalBackbone=ResNet-32, Imbalance Ratio=502021.1176.7
CEBackbone=ResNet-32, Imbalance Ratio=502021.1174.8