Few-shot Classification on ChestX
26.05AccuracyPsCo
Evaluation Results
| Method | Links | |||
|---|---|---|---|---|
| PsCoBackbone=ResNet-50, Base Method=BYOL, Evaluation protocol=5-way 5-shot2023.03 | 26.05 | — | — | |
| CLIPNumber of ways=5, Number of shots=52023.10 | 25.58 | — | — | |
| SupervisedBackbone=ResNet-50, Evaluation protocol=5-way 5-shot2023.03 | 25.35 | — | — | |
| WiSE-FTNumber of ways=5, Number of shots=52023.10 | 25.08 | — | — | |
| PsCoBackbone=ResNet-18, Base Method=MoCo v2, Evaluation protocol=5-way 5-shot2023.03 | 25.03 | — | — | |
| FD-AlignNumber of ways=5, Number of shots=52023.10 | 24.95 | — | — | |
| MoCo v2Backbone=ResNet-18, Evaluation protocol=5-way 5-shot2023.03 | 24.34 | — | — | |
| MoCo v2Backbone=ResNet-50, Evaluation protocol=5-way 5-shot2023.03 | 24.28 | — | — | |
| BYOLBackbone=ResNet-50, Evaluation protocol=5-way 5-shot2023.03 | 24.15 | — | — | |
| PsCoBackbone=ResNet-50, Base Method=MoCo v2, Evaluation protocol=5-way 5-shot2023.03 | 23.6 | — | — | |
| CLIPNumber of ways=5, Number of shots=12023.10 | 22.65 | — | — | |
| FD-AlignNumber of ways=5, Number of shots=12023.10 | 22.31 | — | — | |
| WiSE-FTNumber of ways=5, Number of shots=12023.10 | 22.27 | — | — | |
| MAMLLearning Paradigm=Supervised meta-learning, Backbone=Conv5, Meta-training dataset=miniImageNet, Number of evaluation tasks=20002023.03 | — | 22.61 | 24.25 | |
| Meta-GMVAELearning Paradigm=Unsupervised meta-learning, Backbone=Conv5, Meta-training dataset=miniImageNet, Number of evaluation tasks=20002023.03 | — | 23.23 | 26.26 | |
| Meta-SVEBMLearning Paradigm=Unsupervised meta-learning, Backbone=Conv5, Meta-training dataset=miniImageNet, Number of evaluation tasks=2002023.03 | — | 26.26 | 28.91 | |
| MoCo v2Learning Paradigm=Self-supervised learning, Backbone=Conv5, Meta-training dataset=miniImageNet, Number of evaluation tasks=20002023.03 | — | 25.24 | 29.19 | |
| ProtoNetsLearning Paradigm=Supervised meta-learning, Backbone=Conv5, Meta-training dataset=miniImageNet, Number of evaluation tasks=20002023.03 | — | 23.15 | 25.72 | |
| PsCoLearning Paradigm=Unsupervised meta-learning, Backbone=Conv5, Meta-training dataset=miniImageNet, Number of evaluation tasks=20002023.03 | — | 24.78 | 27.69 | |
| SimCLRLearning Paradigm=Self-supervised learning, Backbone=Conv5, Meta-training dataset=miniImageNet, Number of evaluation tasks=20002023.03 | — | 25.14 | 29.21 | |
| SwAVLearning Paradigm=Self-supervised learning, Backbone=Conv5, Meta-training dataset=miniImageNet, Number of evaluation tasks=20002023.03 | — | 24.99 | 28.57 |