CIFAR-FS Leaderboard (5-class)
Edit this leaderboard
Method | Venue | Year | Backbone | Setting | 1-shot | 5-shot | Code | Reported by |
---|---|---|---|---|---|---|---|---|
FewTURE | NeurIPS | 2022 | Swin-Tiny | Inductive | 77.76 ± 0.81 | 88.90 ± 0.59 | [PyTorch] | [Source] |
FewTURE | NeurIPS | 2022 | ViT-Small | Inductive | 76.10 ± 0.88 | 86.14 ± 0.64 | [PyTorch] | [Source] |
HCTransformers | CVPR | 2022 | ViT-S | Inductive | 78.89 ± 0.18 | 90.50 ± 0.09 | [PyTorch] | [Source] |
EASY | arXiv | 2022 | 3xResNet-12 | Transductive | 87.16 ± 0.21 | 90.47 ± 0.15 | [PyTorch] | [Source] |
EASY | arXiv | 2022 | 2xResNet-12(1/√2) | Transductive | 86.99 ± 0.21 | 90.20 ± 0.15 | [PyTorch] | [Source] |
EASY | arXiv | 2022 | 3xResNet-12 | Inductive | 76.20 ± 0.2 | 89.00 ± 0.14 | [PyTorch] | [Source] |
EASY | arXiv | 2022 | 2xResNet-12(1/√2) | Inductive | 75.24 ± 0.2 | 88.38 ± 0.1 | [PyTorch] | [Source] |
MetaOptNet | CVPR | 2019 | ResNet-12 | Inductive | 72.0 ± 0.7 | 84.2 ± 0.5 | [PyTorch] | [Source] |
Dhillon et al. | ICLR | 2020 | WRN-28-10 | Transductive | 76.58 ± 0.68 | 85.79 ± 0.50 | None | [Source] |
SIB | ICLR | 2020 | WRN-28-10 | Transductive | 80.0 ± 0.6 | 85.3 ± 0.4 | [PyTorch] | [Source] |
SIB | ICLR | 2020 | 4CONV | Transductive | 68.7 ± 0.6 | 77.1 ± 0.4 | [PyTorch] | [Source] |
Ravichandran et al. | ICCV | 2019 | 4CONV | Inductive | 55.14 ± 0.48 | 71.66 ± 0.39 | None | [Source] |
Ravichandran et al. | ICCV | 2019 | ResNet-12 | Inductive | 69.15 | 84.70 | None | [Source] |
TEAM | ICCV | 2019 | ResNet-12 | Transductive | 70.43 | 81.25 | None | [Source] |
TEAM | ICCV | 2019 | 4CONV | Transductive | 64.07 | 79.05 | None | [Source] |
CC+rot | ICCV | 2019 | WRN-28-10 | Semi-supervised | 75.38 ± 0.31 | 87.25 ± 0.21 | [PyTorch] | [Source] |
DSN-MR | CVPR | 2020 | ResNet-12 | Transductive | 75.6 ± 0.9 | 86.2 ± 0.6 | [PyTorch] | [Source] |
DPGN | CVPR | 2020 | ResNet-12 | Transductive | 77.9 ± 0.5 | 90.2 ± 0.4 | [PyTorch] | [Source] |
DPGN | CVPR | 2020 | 4CONV | Transductive | 76.4 ± 0.5 | 88.4 ± 0.4 | [PyTorch] | [Source] |
ICI | CVPR | 2020 | ResNet-12 | Transductive | 73.97 | 84.13 | [PyTorch] | [Source] |
ICI | CVPR | 2020 | ResNet-12 | Semi-supervised | 76.51 | 84.32 | [PyTorch] | [Source] |
J. Kim et al. | ECCV | 2020 | ResNet-12 | Inductive | 73.51 ± 0.92 | 85.65 ± 0.65 | [PyTorch] | [Source] |
Y. Tian et al. | ECCV | 2020 | ResNet-12 | Inductive | 73.9 ± 0.8 | 86.9 ± 0.5 | [PyTorch] | [Source] |
ICI v2 | TPAMI | 2021 | ResNet-12 | Semi-supervised | 80.74 ± 0.61 | 87.16 ± 0.36 | [PyTorch] | [Source] |
ICI v2 | TPAMI | 2021 | ResNet-12 | Transductive | 79.19 ± 0.63 | 86.66 ± 0.36 | [PyTorch] | [Source] |
ConstellationNet | ICLR | 2021 | ResNet-12 | Inductive | 75.4 ± 0.2 | 86.8 ± 0.2 | None | [Source] |
ConstellationNet | ICLR | 2021 | 4CONV | Inductive | 69.3 ± 0.3 | 82.7 ± 0.2 | None | [Source] |
PT+MAP | arXiv | 2021 | WRN | Inductive | 74.64 ± 0.21 | 87.64 ± 0.15 | [PyTorch] | [Source] |
PT+MAP | arXiv | 2021 | WRN | Transductive | 87.69 ± 0.23 | 90.68 ± 0.15 | [PyTorch] | [Source] |
ECKPN | CVPR | 2021 | 4CONV | Transductive | 77.5±0.4 | 89.1±0.5 | None | [Source] |
ECKPN | CVPR | 2021 | ResNet-12 | Transductive | 79.2±0.4 | 91.0±0.5 | None | [Source] |
Rizve et al. | CVPR | 2021 | ResNet-12 | Inductive | 77.87±0.85 | 89.74±0.57 | None | [Source] |
PLCM | ICCV | 2021 | ResNet-12 | Semi-supervised | 77.62±1.15 | 86.13±0.67 | None | [Source] |
Curvature Generation | ICCV | 2021 | ResNet-12 (64-160-320-640) | Inductive | 73.0±0.7 | 85.8±0.5 | [PyTorch] | [Source] |
Curvature Generation | ICCV | 2021 | ResNet-12 (64-160-320-640) | Transductive | 76.8±0.7 | 86.4±0.5 | [PyTorch] | [Source] |
TPMN | ICCV | 2021 | ResNet-12 | Inductive | 75.5±0.9 | 87.2±0.6 | None | [Source] |
RENet | ICCV | 2021 | ResNet-12 | Inductive | 74.51±0.46 | 86.60±0.32 | [PyTorch] | [Source] |
BML | ICCV | 2021 | ResNet-12 | Inductive | 73.45±0.47 | 88.04±0.33 | [PyTorch] | [Source] |
MetaQDA | ICCV | 2021 | WRN-28-10 | Inductive | 75.83±0.88 | 88.79±0.75 | [PyTorch] | [Source] |
Meta Navigator | ICCV | 2021 | ResNet-12 | Inductive | 74.63±0.91 | 86.45±0.59 | None | [Source] |
iLPC | ICCV | 2021 | ResNet-12 | Transductive | 77.14±0.95 | 85.23±0.5 | [PyTorch] | [Source] |
iLPC | ICCV | 2021 | WRN-28-10 | Transductive | 86.51±0.75 | 90.60±0.48 | [PyTorch] | [Source] |
iLPC | ICCV | 2021 | ResNet-12 | Semi-supervised | 78.57±0.80 | 85.84±0.56 | [PyTorch] | [Source] |
iLPC | ICCV | 2021 | WRN-28-10 | Semi-supervised | 90.34±0.50 | 91.69±0.55 | [PyTorch] | [Source] |
AIM | ICCV | 2021 | WRN-28-10 | Transductive | 80.20±0.55 | 87.34±0.36 | [PyTorch] | [Source] |
Wang et al. | NeurIPS | 2021 | 4CONV | Inductive | 64.17±0.31 | 78.42±0.26 | None | [Source] |
NCA nearest centroid | NeurIPS | 2021 | ResNet-12 | Inductive | 72.49±0.12 | 85.15±0.09 | [PyTorch] | [Source] |
SSR | NeurIPS | 2021 | ResNet-12 | Transductive | 76.8±0.6 | 83.7±0.4 | [PyTorch] | [Source] |
SSR | NeurIPS | 2021 | WRN-28-10 | Transductive | 81.6±0.6 | 86.0±0.4 | [PyTorch] | [Source] |