List your company
ResearchFormerly WizWand

Image Classification on USPS -> MNIST (test)

99.62AccuracySupTgt
30.855248.707666.5684.4124Oct 15, 2015Jun 14, 2016Feb 12, 2017Oct 13, 2017Jun 13, 2018Feb 11, 2019Oct 12, 2019
Updated 8mo ago

Evaluation results

MethodLinks
SupTgtTraining Protocol=Train on Target, Augmentation Type=TF2017.0699.62
SupTgtTraining Protocol=Train on Target, Augmentation Type=TFA2017.0699.62
Target only (baseline)training=Target domain only2019.1099.6
Mean TeacherTraining Protocol=Unsupervised Domain Adaptation, Augmentation Type=TFA, Confidence Thresholding=true2017.0699.54
SupTgtTraining Protocol=Train on Target, Augmentation Type=minimal2017.0699.53
Target Supervisedtraining=full supervised2019.0999.2
DTA2019.1099.1
3CATN2019.0998.3
Mean TeacherTraining Protocol=Unsupervised Domain Adaptation, Augmentation Type=TF2017.0698.07
CDANvariant=CDAN+E2019.0998
SBADA-GANTraining Protocol=Unsupervised Domain Adaptation2017.0697.6
SE*Protocol=MT+CT+TF2019.1097.3
Mean TeacherTraining Protocol=Unsupervised Domain Adaptation, Augmentation Type=TF, Confidence Thresholding=true2017.0697.28
CyCADA2019.0996.5
MCD (Ours)n (number of feature generator updates)=42017.1294.1
UNIT2019.0993.6
Mean TeacherTraining Protocol=Unsupervised Domain Adaptation, Augmentation Type=minimal, Confidence Thresholding=true2017.0692.35
SupSrcTraining Protocol=Train on Source, Augmentation Type=TFA2017.0691.97
MCD (Ours)n (number of feature generator updates)=32017.1291.8
G2ATraining Protocol=Unsupervised Domain Adaptation2017.0690.8
Proposed MethodTask=Rotation2019.0990.2
ADDA2017.0290.1
ADDA2017.1290.1
ADDATraining Protocol=Unsupervised Domain Adaptation2017.0690.1
ADDA2019.0990.1
ADDA2019.0990.1
MCD (Ours)n (number of feature generator updates)=22017.1290
RevGrad2019.0989.9
CoGAN2017.0289.1
CoGAN2019.0989.1
Source only (Our method)mode=source only2019.0981.4
SupSrcTraining Protocol=Train on Source, Augmentation Type=minimal2017.0677.55
SupSrcTraining Protocol=Train on Source, Augmentation Type=TF2017.0677.53
Source only (baseline)training=Source domain only2019.1076.9
RevGradTraining Protocol=Unsupervised Domain Adaptation2017.0674.01
DRCN2017.1273.7
DRCN2019.0973.7
DCRNTraining Protocol=Unsupervised Domain Adaptation2017.0673.67
Gradient reversal2017.0273
DANNlabeled target samples for validation=true2017.1273
Source onlytraining=source only2019.0969.6
JGSAkernel=primal2017.0568.15
Domain confusion2017.0266.5
Source Onlyadaptation=none2017.1263.4
JDA2017.0559.65
Source only2017.0257.1
TJMBase Classifier=1NN2015.1052.65
TJM2017.0552.25
TCA2017.0551.2
SA2017.0548.8
SCABase Classifier=1NN, Supervision=Semi-supervised2015.1048
SCA2017.0548
GFK2017.0546.45
uSCABase Classifier=1NN, Supervision=Unsupervised2015.1044.86
Rawadaptation=none2017.0544.7
KPCABase Classifier=1NN2015.1042.55
TCABase Classifier=1NN2015.1041.75
SABase Classifier=1NN2015.1041.5
TSCBase Classifier=1NN2015.1040.95
SSTCABase Classifier=1NN2015.1040.07
SDA2017.0535.7
RawBase Classifier=1NN2015.1034.8
GFKBase Classifier=1NN2015.1033.5