Digit Classification on SVHN -> MNIST (test)
99.2AccuracyTarget Supervised
Evaluation Results
| Method | Links | |
|---|---|---|
| Target Supervisedtraining=full supervised2019.09 | 99.2 | |
| target onlyadaptation=none, train_domain=target2018.03 | 98.7 | |
| DeepJDOTadaptation=source to target2018.03 | 96.7 | |
| Fully SupervisedArchitecture=Simple CNN (2 conv, 2 fc, BN)2021.12 | 96 | |
| AssocDAadaptation=source to target2018.03 | 95.7 | |
| IntraDAArchitecture=LeNet variant2020.04 | 95.1 | |
| PFAN2018.11 | 93.9 | |
| Self-ensembleadaptation=source to target2018.03 | 93.33 | |
| 3CATN2019.09 | 92.5 | |
| GenToAdaptadaptation=source to target2018.03 | 92.4 | |
| MSTN2018.11 | 91.7 | |
| UNITnote=uses a larger training set2019.09 | 90.5 | |
| UNITadaptation=source to target2018.03 | 90.5 | |
| CyCADAArchitecture=LeNet variant2020.04 | 90.4 | |
| CDANvariant=CDAN+E2019.09 | 89.2 | |
| ATT2018.11 | 85 | |
| DSNadaptation=source to target2018.03 | 82.7 | |
| DRCN2019.09 | 82 | |
| DRCNadaptation=source to target2018.03 | 81.9 | |
| I2I Adaptadaptation=source to target2018.03 | 80.3 | |
| ADDAadaptation=source to target2018.03 | 76.05 | |
| ADDA2018.11 | 76 | |
| ADDA2019.09 | 76 | |
| ADDAArchitecture=LeNet variant2020.04 | 76 | |
| DeepJDOTadaptation=source to target, eval_domain=source2018.03 | 75.7 | |
| RevGrad2018.11 | 73.9 | |
| DANN2019.09 | 73.6 | |
| DANNadaptation=source to target2018.03 | 70.8 | |
| DUAArchitecture=Simple CNN (2 conv, 2 fc, BN)2021.12 | 68 | |
| StochJDOTadaptation=source to target2018.03 | 67.6 | |
| Source onlytraining=source only2019.09 | 67.1 | |
| Source onlyArchitecture=LeNet variant2020.04 | 67.1 | |
| MMDadaptation=source to target2018.03 | 64.8 | |
| Source onlyadaptation=none, train_domain=source2018.03 | 60.7 | |
| Source Only2018.11 | 60.1 | |
| DeepCORALadaptation=source to target2018.03 | 59.6 | |
| Source onlyArchitecture=Simple CNN (2 conv, 2 fc, BN)2021.12 | 58 |