Few-shot Classification on MiniImageNet (meta-test)
33.82Best Accuracy (5-way 1-shot)TDFML-ProtoNet (+mixup)
Evaluation results
| Method | Links | ||||||
|---|---|---|---|---|---|---|---|
| TDFML-ProtoNet (+mixup)Base Learner=ProtoNet, Augmentation=mixup2023.11 | 33.82 | 32.75 | 43.97 | 42.88 | — | — | |
| TDFML-ProtoNetBase Learner=ProtoNet2023.11 | 32.61 | 31.97 | 42.93 | 41.28 | — | — | |
| TDFML-ANILBase Learner=ANIL2023.11 | 32.58 | 29.76 | 43.63 | 42.45 | — | — | |
| PURER-ProtoNetBase Learner=ProtoNet2023.11 | 30.46 | 24 | 41 | 31.32 | — | — | |
| PURER-ANILBase Learner=ANIL2023.11 | 30.2 | 23.05 | 40.78 | 29.6 | — | — | |
| DFL2L2023.11 | 27.56 | 25.22 | 30.19 | 28.43 | — | — | |
| RANDOM2023.11 | 25.06 | 25.06 | 28.1 | 28.1 | — | — | |
| AVERAGE2023.11 | 23.79 | 23.79 | 27.49 | 27.49 | — | — | |
| Best pre-trained model2026.07 | — | — | — | — | 24.5 | 31.6 | |
| FREE2026.07 | — | — | — | — | 25.2 | 35.8 | |
| K-nearest neighbors (embedded)2026.07 | — | — | — | — | 28.8 | 39.2 | |
| K-nearest neighbors (pixel)2026.07 | — | — | — | — | 27.7 | 30.4 | |
| KD-MAML2026.07 | — | — | — | — | 28.3 | 36.3 | |
| Ours2026.07 | — | — | — | — | 33.6 | 48.7 | |
| PsCo2026.07 | — | — | — | — | 31.7 | 38.5 | |
| Random2026.07 | — | — | — | — | 28.1 | 37.2 |