Few-shot Classification on ISIC (5-way 5-shot and 20-shot)
47.34Accuracy (5-way 5-shot)MAML
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| MAMLLearning Paradigm=Supervised meta-learning, Backbone=Conv5, Meta-training dataset=miniImageNet, Number of evaluation tasks=20002023.03 | 47.34 | 55.09 | |
| PsCoLearning Paradigm=Unsupervised meta-learning, Backbone=Conv5, Meta-training dataset=miniImageNet, Number of evaluation tasks=20002023.03 | 44 | 54.59 | |
| MoCo v2Learning Paradigm=Self-supervised learning, Backbone=Conv5, Meta-training dataset=miniImageNet, Number of evaluation tasks=20002023.03 | 43.43 | 52.14 | |
| SwAVLearning Paradigm=Self-supervised learning, Backbone=Conv5, Meta-training dataset=miniImageNet, Number of evaluation tasks=20002023.03 | 43.21 | 51.99 | |
| SimCLRLearning Paradigm=Self-supervised learning, Backbone=Conv5, Meta-training dataset=miniImageNet, Number of evaluation tasks=20002023.03 | 42.83 | 51.35 | |
| ProtoNetsLearning Paradigm=Supervised meta-learning, Backbone=Conv5, Meta-training dataset=miniImageNet, Number of evaluation tasks=20002023.03 | 40.62 | 48.38 | |
| Meta-SVEBMLearning Paradigm=Unsupervised meta-learning, Backbone=Conv5, Meta-training dataset=miniImageNet, Number of evaluation tasks=2002023.03 | 38.85 | 48.43 | |
| Meta-GMVAELearning Paradigm=Unsupervised meta-learning, Backbone=Conv5, Meta-training dataset=miniImageNet, Number of evaluation tasks=20002023.03 | 33.48 | 39.48 |