Image Classification on USPS -> MNIST (test)
99.62AccuracySupTgt
Evaluation Results
| Method | Links | |
|---|---|---|
| SupTgtTraining Protocol=Train on Target, Augmentation Type=TF2017.06 | 99.62 | |
| SupTgtTraining Protocol=Train on Target, Augmentation Type=TFA2017.06 | 99.62 | |
| Target only (baseline)training=Target domain only2019.10 | 99.6 | |
| Mean TeacherTraining Protocol=Unsupervised Domain Adaptation, Augmentation Type=TFA, Confidence Thresholding=true2017.06 | 99.54 | |
| SupTgtTraining Protocol=Train on Target, Augmentation Type=minimal2017.06 | 99.53 | |
| Target Supervisedtraining=full supervised2019.09 | 99.2 | |
| DTA2019.10 | 99.1 | |
| 3CATN2019.09 | 98.3 | |
| Mean TeacherTraining Protocol=Unsupervised Domain Adaptation, Augmentation Type=TF2017.06 | 98.07 | |
| CDANvariant=CDAN+E2019.09 | 98 | |
| SBADA-GANTraining Protocol=Unsupervised Domain Adaptation2017.06 | 97.6 | |
| SE*Protocol=MT+CT+TF2019.10 | 97.3 | |
| Mean TeacherTraining Protocol=Unsupervised Domain Adaptation, Augmentation Type=TF, Confidence Thresholding=true2017.06 | 97.28 | |
| CyCADA2019.09 | 96.5 | |
| MCD (Ours)n (number of feature generator updates)=42017.12 | 94.1 | |
| UNIT2019.09 | 93.6 | |
| Mean TeacherTraining Protocol=Unsupervised Domain Adaptation, Augmentation Type=minimal, Confidence Thresholding=true2017.06 | 92.35 | |
| SupSrcTraining Protocol=Train on Source, Augmentation Type=TFA2017.06 | 91.97 | |
| MCD (Ours)n (number of feature generator updates)=32017.12 | 91.8 | |
| G2ATraining Protocol=Unsupervised Domain Adaptation2017.06 | 90.8 | |
| Proposed MethodTask=Rotation2019.09 | 90.2 | |
| ADDA2017.02 | 90.1 | |
| ADDA2017.12 | 90.1 | |
| ADDATraining Protocol=Unsupervised Domain Adaptation2017.06 | 90.1 | |
| ADDA2019.09 | 90.1 | |
| ADDA2019.09 | 90.1 | |
| MCD (Ours)n (number of feature generator updates)=22017.12 | 90 | |
| RevGrad2019.09 | 89.9 | |
| CoGAN2017.02 | 89.1 | |
| CoGAN2019.09 | 89.1 | |
| Source only (Our method)mode=source only2019.09 | 81.4 | |
| SupSrcTraining Protocol=Train on Source, Augmentation Type=minimal2017.06 | 77.55 | |
| SupSrcTraining Protocol=Train on Source, Augmentation Type=TF2017.06 | 77.53 | |
| Source only (baseline)training=Source domain only2019.10 | 76.9 | |
| RevGradTraining Protocol=Unsupervised Domain Adaptation2017.06 | 74.01 | |
| DRCN2017.12 | 73.7 | |
| DRCN2019.09 | 73.7 | |
| DCRNTraining Protocol=Unsupervised Domain Adaptation2017.06 | 73.67 | |
| Gradient reversal2017.02 | 73 | |
| DANNlabeled target samples for validation=true2017.12 | 73 | |
| Source onlytraining=source only2019.09 | 69.6 | |
| JGSAkernel=primal2017.05 | 68.15 | |
| Domain confusion2017.02 | 66.5 | |
| Source Onlyadaptation=none2017.12 | 63.4 | |
| JDA2017.05 | 59.65 | |
| Source only2017.02 | 57.1 | |
| TJMBase Classifier=1NN2015.10 | 52.65 | |
| TJM2017.05 | 52.25 | |
| TCA2017.05 | 51.2 | |
| SA2017.05 | 48.8 | |
| SCABase Classifier=1NN, Supervision=Semi-supervised2015.10 | 48 | |
| SCA2017.05 | 48 | |
| GFK2017.05 | 46.45 | |
| uSCABase Classifier=1NN, Supervision=Unsupervised2015.10 | 44.86 | |
| Rawadaptation=none2017.05 | 44.7 | |
| KPCABase Classifier=1NN2015.10 | 42.55 | |
| TCABase Classifier=1NN2015.10 | 41.75 | |
| SABase Classifier=1NN2015.10 | 41.5 | |
| TSCBase Classifier=1NN2015.10 | 40.95 | |
| SSTCABase Classifier=1NN2015.10 | 40.07 | |
| SDA2017.05 | 35.7 | |
| RawBase Classifier=1NN2015.10 | 34.8 | |
| GFKBase Classifier=1NN2015.10 | 33.5 |