List your company
ResearchFormerly WizWand

Image Classification on CIFAR-10 standard (test) (Top-1 Error Rate)

4.08Top-1 Error RateDash (RA)
2.072815.621429.1742.7186Sep 1, 2021
Updated 8mo ago

Evaluation results

MethodLinks
Dash (RA)Number of labels=4000, Backbone=Wide ResNet-28-22021.094.08
Dash (CTA)Number of labels=4000, Backbone=Wide ResNet-28-22021.094.13
FixMatch (RA)Number of labels=4000, Backbone=Wide ResNet-28-22021.094.26
FixMatch (CTA)Number of labels=4000, Backbone=Wide ResNet-28-22021.094.31
RYS (FixMatch)Number of labels=4000, Backbone=Wide ResNet-28-22021.094.35
Dash (RA)Number of labels=250, Backbone=Wide ResNet-28-22021.094.56
ReMixMatchNumber of labels=4000, Backbone=Wide ResNet-28-22021.094.72
RYS (UDA)Number of labels=4000, Backbone=Wide ResNet-28-22021.094.75
Dash (CTA)Number of labels=250, Backbone=Wide ResNet-28-22021.094.78
UDANumber of labels=4000, Backbone=Wide ResNet-28-22021.094.88
RYS (FixMatch)Number of labels=250, Backbone=Wide ResNet-28-22021.095.05
FixMatch (CTA)Number of labels=250, Backbone=Wide ResNet-28-22021.095.07
FixMatch (RA)Number of labels=250, Backbone=Wide ResNet-28-22021.095.07
ReMixMatchNumber of labels=250, Backbone=Wide ResNet-28-22021.095.44
RYS (UDA)Number of labels=250, Backbone=Wide ResNet-28-22021.095.53
MixMatchNumber of labels=4000, Backbone=Wide ResNet-28-22021.096.42
UDANumber of labels=250, Backbone=Wide ResNet-28-22021.098.82
Dash (CTA)Number of labels=40, Backbone=Wide ResNet-28-22021.099.16
Mean TeacherNumber of labels=4000, Backbone=Wide ResNet-28-22021.099.19
MixMatchNumber of labels=250, Backbone=Wide ResNet-28-22021.0911.05
FixMatch (CTA)Number of labels=40, Backbone=Wide ResNet-28-22021.0911.39
Dash (RA)Number of labels=40, Backbone=Wide ResNet-28-22021.0913.22
FixMatch (RA)Number of labels=40, Backbone=Wide ResNet-28-22021.0913.81
Π-ModelNumber of labels=4000, Backbone=Wide ResNet-28-22021.0914.01
Pseudo-LabelingNumber of labels=4000, Backbone=Wide ResNet-28-22021.0916.09
ReMixMatchNumber of labels=40, Backbone=Wide ResNet-28-22021.0919.1
UDANumber of labels=40, Backbone=Wide ResNet-28-22021.0929.05
Mean TeacherNumber of labels=250, Backbone=Wide ResNet-28-22021.0932.32
MixMatchNumber of labels=40, Backbone=Wide ResNet-28-22021.0947.54
Pseudo-LabelingNumber of labels=250, Backbone=Wide ResNet-28-22021.0949.78
Π-ModelNumber of labels=250, Backbone=Wide ResNet-28-22021.0954.26