Image Classification on CIFAR-10 400 labels per class (test)
88.44AccuracyFlowGMM Sup
Evaluation results
| Method | Links | |
|---|---|---|
| FlowGMM SupTraining Labels=All labels, Learning Paradigm=Supervised2019.12 | 88.44 | |
| Π-ModelLearning Paradigm=Semi-supervised2019.12 | 87.64 | |
| BadGANLearning Paradigm=Semi-supervised2019.12 | 85.59 | |
| CLS-GANlabeled_examples_per_class=4002017.01 | 82.7 | |
| ALIlabeled_examples_per_class=4002017.01 | 81.7 | |
| Improved GANlabeled_examples_per_class=4002017.01 | 81.4 | |
| FlowGMM-consLearning Paradigm=Semi-supervised, Consistency Regularization=true2019.12 | 80.9 | |
| CatGANlabeled_examples_per_class=4002017.01 | 80.4 | |
| Ladder Networklabeled_examples_per_class=4002017.01 | 79.6 | |
| FlowGMMLearning Paradigm=Semi-supervised2019.12 | 78.24 | |
| Exemplar CNNmax # of features units=1024, Classifier=Linear L2-SVM2015.11 | 77.4 | |
| Examplar CNNlabeled_examples_per_class=4002017.01 | 77.4 | |
| Conditional GANlabeled_examples_per_class=4002017.01 | 75.5 | |
| DCGAN + L2-SVMmax # of features units=512, Classifier=Linear L2-SVM, Pre-trained on=ImageNet-1k2015.11 | 73.8 | |
| DCGANlabeled_examples_per_class=4002017.01 | 73.8 | |
| FlowGMM SupTraining Labels=ni labels, Learning Paradigm=Supervised2019.12 | 73.13 | |
| View Invariant K-meansmax # of features units=6400, Classifier=Linear L2-SVM2015.11 | 72.6 | |
| View Invariant K-meanslabeled_examples_per_class=4002017.01 | 72.6 | |
| 3 Layer K-means Learned RFmax # of features units=3200, Classifier=Linear L2-SVM2015.11 | 70.7 | |
| 3 Layer K-means Learned RFlabeled_examples_per_class=4002017.01 | 70.7 | |
| 1 Layer K-meansmax # of features units=4800, Classifier=Linear L2-SVM2015.11 | 63.7 | |
| 1 Layer K-meanslabeled_examples_per_class=4002017.01 | 63.7 |