Image Classification on CIFAR10 10-way 5-shot
73.8AccuracyFAAST
Evaluation results
| Method | Links | |
|---|---|---|
| FAASTAdaptation=Fast weights, Inference Compute=O(d_x^2), Inference Memory=4 MB, Learn Cost GPU Secs=2.1 s (-94.7%)2026.05 | 73.8 | |
| Linear projectionAdaptation=Backprop, Inference Compute=O(d_x^2), Inference Memory=4 MB, Learn Cost GPU Secs=780 s2026.05 | 71.9 | |
| k-NN MemoryAdaptation=Memory, Inference Compute=O(N d_x), Inference Memory=80 MB, Learn Cost GPU Secs=41 s2026.05 | 71.7 | |
| Softmax MemoryAdaptation=Memory, Inference Compute=O(N d_x), Inference Memory=80 MB2026.05 | 71.5 | |
| CLIPAdaptation=No adapt, Inference Compute=O(d_x), Inference Memory=0 MB2026.05 | 70.3 | |
| Full FinetuningAdaptation=Backprop, Inference Compute=O(d_x^2), Inference Memory=4 MB, Learn Cost GPU Secs=1,603 s2026.05 | 13.3 |