Image Classification on STL10 scale (test)
73.32AccuracyScale-Equivariant Fourier Networks
Evaluation results
| Method | Links | |||
|---|---|---|---|---|
| Scale-Equivariant Fourier NetworksArchitecture=six scale-equivariant Fourier blocks followed by a two-layer MLP2023.11 | 73.32 | 67.7 | 0 | |
| Fourier CNNArchitecture=six Fourier blocks followed by a two-layered MLP2023.11 | 58.44 | 28.12 | 0.19 | |
| Wide ResNetbaseline=CNN baseline2023.11 | 55.96 | 29.16 | 0.16 | |
| SI-CovNet2023.11 | 55.88 | 21.87 | 0.03 | |
| SESN2023.11 | 55.25 | 41.66 | 0.04 | |
| DSS2023.11 | 53.47 | 19.79 | 0.02 | |
| SS-CNN2023.11 | 47.88 | 19.79 | 1.82 | |
| DISCO2023.11 | 47.68 | 35.41 | 0.06 |