Few Shot Image Classification On Omniglot 5 2
評価指標
Accuracy
評価結果
このベンチマークにおける各モデルのパフォーマンス結果
モデル名 | Accuracy | Paper Title | Repository |
|---|---|---|---|
| Relation Net | 99.8 | Learning to Compare: Relation Network for Few-Shot Learning | |
| Prototypical Networks | 99.7 | Prototypical Networks for Few-shot Learning | |
| VAMPIRE | 99.56% | Uncertainty in Model-Agnostic Meta-Learning using Variational Inference | |
| MAML++ | 99.85% | How to train your MAML | |
| MC2+ | 99.89 | Meta-Curvature | |
| adaCNN (DF) | 99.37 | Rapid Adaptation with Conditionally Shifted Neurons | - |
| iMAML, Hessian-Free | 99.74% | Meta-Learning with Implicit Gradients | |
| Neural Statistician | 99.5 | Towards a Neural Statistician | |
| DCN6-E | 99.92% | Decoder Choice Network for Meta-Learning | |
| DCN4 | 99.89% | Decoder Choice Network for Meta-Learning | |
| Hyperbolic ProtoNet | 99.4 | Hyperbolic Image Embeddings | |
| Matching Nets | 98.9 | Matching Networks for One Shot Learning | |
| ConvNet with Memory Module | 99.6 | Learning to Remember Rare Events | |
| MAML | 99.9 | Model-Agnostic Meta-Learning for Fast Adaptation of Deep Networks | |
| APL | 99.9 | Adaptive Posterior Learning: few-shot learning with a surprise-based memory module | |
| Reptile + Transduction | 99.48 | On First-Order Meta-Learning Algorithms |
0 of 16 row(s) selected.