List your company
ResearchFormerly WizWand

Image Classification on CIFAR-10-LT (test)

0.0824Top-1 ErrorLDAM-DRW + Du @5x
0.0779120.1082060.13850.168794Jun 13, 2020Oct 21, 2020Mar 1, 2021Jul 9, 2021Nov 17, 2021Mar 27, 2022Aug 5, 2022
Updated 8mo ago

Evaluation results

MethodLinks
LDAM-DRW + Du @5xLabeled Imbalance Ratio (p)=10, Unlabeled Imbalance Ratio (pu)=p/2, Backbone=ResNet-32, Amount of Unlabeled Data=5x2020.060.0824—
LDAM-DRW + Du @5xLabeled Imbalance Ratio (p)=10, Unlabeled Imbalance Ratio (pu)=p, Backbone=ResNet-32, Amount of Unlabeled Data=5x2020.060.0868—
LDAM-DRW + Du @5xLabeled Imbalance Ratio (p)=10, Unlabeled Imbalance Ratio (pu)=1, Backbone=ResNet-32, Amount of Unlabeled Data=5x2020.060.0872—
LDAM-DRW + Du @5xLabeled Imbalance Ratio (p)=10, Unlabeled Imbalance Ratio (pu)=2p, Backbone=ResNet-32, Amount of Unlabeled Data=5x2020.060.0897—
CE + Du @5xLabeled Imbalance Ratio (p)=10, Unlabeled Imbalance Ratio (pu)=1, Backbone=ResNet-32, Amount of Unlabeled Data=5x2020.060.1022—
MetaSAug with LDAM lossImbalance factor=10, Backbone=ResNet-322021.030.1032—
CE + Du @5xLabeled Imbalance Ratio (p)=10, Unlabeled Imbalance Ratio (pu)=p/2, Backbone=ResNet-32, Amount of Unlabeled Data=5x2020.060.1048—
MetaSAug with cross-entropy lossImbalance factor=10, Backbone=ResNet-322021.030.1056—
MetaSAug with focal lossImbalance factor=10, Backbone=ResNet-322021.030.1074—
CE + Du @5xLabeled Imbalance Ratio (p)=10, Unlabeled Imbalance Ratio (pu)=p, Backbone=ResNet-32, Amount of Unlabeled Data=5x2020.060.1086—
CE + Du @5xLabeled Imbalance Ratio (p)=10, Unlabeled Imbalance Ratio (pu)=2p, Backbone=ResNet-32, Amount of Unlabeled Data=5x2020.060.1104—
Meta-class-weight with cross-entropy lossImbalance factor=10, Backbone=ResNet-322021.030.1115—
Meta-class-weight with focal lossImbalance factor=10, Backbone=ResNet-322021.030.1163—
LDAM-DRWImbalance factor=10, Backbone=ResNet-322021.030.1163—
BBNImbalance factor=10, Backbone=ResNet-322021.030.1168—
LDAM-DRWLabeled Imbalance Ratio (p)=10, Backbone=ResNet-32, Amount of Unlabeled Data=None2020.060.1184—
MetaSAug with LDAM lossImbalance factor=20, Backbone=ResNet-322021.030.119—
MetaSAug with LDAM lossBackbone=ResNet-32, Imbalance Factor=202022.080.119—
MetaSAug with cross-entropy lossImbalance factor=20, Backbone=ResNet-322021.030.1236—
MetaSAug with CE lossBackbone=ResNet-32, Imbalance Factor=202022.080.1236—
Meta-weight netImbalance factor=10, Backbone=ResNet-322021.030.1245—
Optimal Transport Re-weighting (Weight Vector)Backbone=ResNet-32, Imbalance Factor=202022.080.125—
Class-balanced focal lossImbalance factor=10, Backbone=ResNet-322021.030.1252—
Meta-class-weight with LDAM lossImbalance factor=10, Backbone=ResNet-322021.030.126—
LDAM lossImbalance factor=10, Backbone=ResNet-322021.030.1268—
MetaSAug with focal lossImbalance factor=20, Backbone=ResNet-322021.030.1284—
MetaSAug with focal lossBackbone=ResNet-32, Imbalance Factor=202022.080.1284—
MixupImbalance factor=10, Backbone=ResNet-322021.030.129—
Class-balanced cross-entropy lossImbalance factor=10, Backbone=ResNet-322021.030.131—
Focal lossImbalance factor=10, Backbone=ResNet-322021.030.1334—
Meta-class-weight with cross-entropy lossImbalance factor=20, Backbone=ResNet-322021.030.1354—
Meta-class-weight with CE lossBackbone=ResNet-32, Imbalance Factor=202022.080.1354—
CELabeled Imbalance Ratio (p)=10, Backbone=ResNet-32, Amount of Unlabeled Data=None2020.060.1361—
Cross-entropy trainingImbalance factor=10, Backbone=ResNet-322021.030.1382—
Meta-class-weight with focal lossImbalance factor=20, Backbone=ResNet-322021.030.139—
Meta-class-weight with focal lossBackbone=ResNet-32, Imbalance Factor=202022.080.139—
IB + Focal lossBackbone=ResNet-32, Imbalance Factor=202022.080.1432—
LDAM-DRW + Du @5xLabeled Imbalance Ratio (p)=50, Unlabeled Imbalance Ratio (pu)=1, Backbone=ResNet-32, Amount of Unlabeled Data=5x2020.060.1433—
IBBackbone=ResNet-32, Imbalance Factor=202022.080.1459—
LDAM-DRW + Du @5xLabeled Imbalance Ratio (p)=50, Unlabeled Imbalance Ratio (pu)=p/2, Backbone=ResNet-32, Amount of Unlabeled Data=5x2020.060.147—
IB + CBBackbone=ResNet-32, Imbalance Factor=202022.080.1473—
LDAM-DRW + Du @5xLabeled Imbalance Ratio (p)=50, Unlabeled Imbalance Ratio (pu)=p, Backbone=ResNet-32, Amount of Unlabeled Data=5x2020.060.1493—
LDAM-DRW + Du @5xLabeled Imbalance Ratio (p)=100, Unlabeled Imbalance Ratio (pu)=1, Backbone=ResNet-32, Amount of Unlabeled Data=5x2020.060.1496—
LDAM-DRWImbalance factor=20, Backbone=ResNet-322021.030.151—
LDAM-DRWBackbone=ResNet-32, Imbalance Factor=202022.080.151—
LDAM-DRW + Du @5xLabeled Imbalance Ratio (p)=100, Unlabeled Imbalance Ratio (pu)=p/2, Backbone=ResNet-32, Amount of Unlabeled Data=5x2020.060.1518—
LDAM-DRW + Du @5xLabeled Imbalance Ratio (p)=50, Unlabeled Imbalance Ratio (pu)=2p, Backbone=ResNet-32, Amount of Unlabeled Data=5x2020.060.1524—
LDAM-DRW + Du @5xLabeled Imbalance Ratio (p)=100, Unlabeled Imbalance Ratio (pu)=p, Backbone=ResNet-32, Amount of Unlabeled Data=5x2020.060.1533—
Class-balanced cross-entropy lossImbalance factor=20, Backbone=ResNet-322021.030.1544—
CB, CE lossBackbone=ResNet-32, Imbalance Factor=202022.080.1544—
Optimal Transport Re-weighting (Weight Vector)Backbone=ResNet-32, Imbalance Factor=502022.080.1554—
LDAM-DRW + Du @5xLabeled Imbalance Ratio (p)=100, Unlabeled Imbalance Ratio (pu)=2p, Backbone=ResNet-32, Amount of Unlabeled Data=5x2020.060.1555—
Meta-weight netImbalance factor=20, Backbone=ResNet-322021.030.1555—
Meta-weight netBackbone=ResNet-32, Imbalance Factor=202022.080.1555—
Meta-class-weight with LDAM lossImbalance factor=20, Backbone=ResNet-322021.030.1563—
Meta-class-weight with LDAM lossBackbone=ResNet-32, Imbalance Factor=202022.080.1563—
MetaSAug with LDAM lossImbalance factor=50, Backbone=ResNet-322021.030.1566—
MetaSAug with LDAM lossBackbone=ResNet-32, Imbalance Factor=502022.080.1566—
MetaSAug with focal lossImbalance factor=50, Backbone=ResNet-322021.030.1596—
MetaSAug with focal lossBackbone=ResNet-32, Imbalance Factor=502022.080.1596—
MetaSAug with cross-entropy lossImbalance factor=50, Backbone=ResNet-322021.030.1597—
MetaSAug with CE lossBackbone=ResNet-32, Imbalance Factor=502022.080.1597—
LDAM lossImbalance factor=20, Backbone=ResNet-322021.030.1611—
LDAM lossBackbone=ResNet-32, Imbalance Factor=202022.080.1611—
Class-balanced focal lossImbalance factor=20, Backbone=ResNet-322021.030.1622—
CB, Focal lossBackbone=ResNet-32, Imbalance Factor=202022.080.1622—
Class-balanced fine-tuningImbalance factor=20, Backbone=ResNet-322021.030.1678—
CE + Du @5xLabeled Imbalance Ratio (p)=50, Unlabeled Imbalance Ratio (pu)=1, Backbone=ResNet-32, Amount of Unlabeled Data=5x2020.060.1679—
Class-balanced fine-tuningImbalance factor=10, Backbone=ResNet-322021.030.1683—
CE + Du @5xLabeled Imbalance Ratio (p)=50, Unlabeled Imbalance Ratio (pu)=p/2, Backbone=ResNet-32, Amount of Unlabeled Data=5x2020.060.1688—
L2RWBackbone=ResNet-32, Imbalance Factor=202022.080.169—
Meta-class-weight with focal lossImbalance factor=50, Backbone=ResNet-322021.030.1712—
Meta-class-weight with focal lossBackbone=ResNet-32, Imbalance Factor=502022.080.1712—
Focal lossImbalance factor=20, Backbone=ResNet-322021.030.1724—
Focal lossBackbone=ResNet-32, Imbalance Factor=202022.080.1724—
CE + Du @5xLabeled Imbalance Ratio (p)=100, Unlabeled Imbalance Ratio (pu)=1, Backbone=ResNet-32, Amount of Unlabeled Data=5x2020.060.1748—
Cross-entropy trainingImbalance factor=20, Backbone=ResNet-322021.030.1756—
CE lossBackbone=ResNet-32, Imbalance Factor=202022.080.1756—
IB + Focal lossBackbone=ResNet-32, Imbalance Factor=502022.080.1762—
Meta-class-weight with LDAM lossImbalance factor=50, Backbone=ResNet-322021.030.1777—
Meta-class-weight with LDAM lossBackbone=ResNet-32, Imbalance Factor=502022.080.1777—
BBNImbalance factor=50, Backbone=ResNet-322021.030.1782—
BBNBackbone=ResNet-32, Imbalance Factor=202022.080.1782—
L2RWImbalance factor=10, Backbone=ResNet-322021.030.1788—
IB + CBBackbone=ResNet-32, Imbalance Factor=502022.080.1791—
Optimal Transport Re-weighting (Weight Vector)Backbone=ResNet-32, Imbalance Factor=1002022.080.1813—
IBBackbone=ResNet-32, Imbalance Factor=502022.080.1834—
CE + Du @5xLabeled Imbalance Ratio (p)=50, Unlabeled Imbalance Ratio (pu)=p, Backbone=ResNet-32, Amount of Unlabeled Data=5x2020.060.1836—
CE + Du @5xLabeled Imbalance Ratio (p)=100, Unlabeled Imbalance Ratio (pu)=p/2, Backbone=ResNet-32, Amount of Unlabeled Data=5x2020.060.1842—
L2RWImbalance factor=20, Backbone=ResNet-322021.030.1865—
LDAM-DRWImbalance factor=50, Backbone=ResNet-322021.030.1873—
LDAM-DRWBackbone=ResNet-32, Imbalance Factor=502022.080.1873—
CE + Du @5xLabeled Imbalance Ratio (p)=100, Unlabeled Imbalance Ratio (pu)=p, Backbone=ResNet-32, Amount of Unlabeled Data=5x2020.060.1874—
LDAM-DRWLabeled Imbalance Ratio (p)=50, Backbone=ResNet-32, Amount of Unlabeled Data=None2020.060.1906—
MetaSAug with LDAM lossImbalance factor=100, Backbone=ResNet-322021.030.1934—
MetaSAug with LDAM lossBackbone=ResNet-32, Imbalance Factor=1002022.080.1934—
MetaSAug with focal lossImbalance factor=100, Backbone=ResNet-322021.030.1936—
MetaSAug with focal lossBackbone=ResNet-32, Imbalance Factor=1002022.080.1936—
MetaSAug with cross-entropy lossImbalance factor=100, Backbone=ResNet-322021.030.1946—
MetaSAug with CE lossBackbone=ResNet-32, Imbalance Factor=1002022.080.1946—