Semi-supervised Classification on ImageNet 10% labels
77.7Top-1 AccBYOL
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| BYOLArch.=R200 (x2), Param.=250M, Repr.=40962021.05 | 77.7 | 93.7 | |
| VICRegArch.=R200 (x2), Param.=250M, Repr.=40962021.05 | 77.3 | 93.6 | |
| BYOLArch.=R50 (x4), Param.=375M, Repr.=81922021.05 | 75.7 | 92.5 | |
| SimCLRArch.=R50 (x4), Param.=375M, Repr.=81922021.05 | 74.4 | 92.6 | |
| UniMatch V2Labeled data ratio=10%2024.10 | 74.3 | 91.7 | |
| SimMatchLabeled data ratio=10%2024.10 | 74.1 | 91.5 | |
| VICRegArch.=R50 (x2), Param.=93M, Repr.=40962021.05 | 73.9 | 91.8 | |
| CoMatchLabeled data ratio=10%2024.10 | 73.6 | 91.6 | |
| BYOLArch.=R50 (x2), Param.=93M, Repr.=40962021.05 | 73.5 | 91.7 | |
| SimCLRArch.=R50 (x2), Param.=93M, Repr.=40962021.05 | 71.7 | 91.2 |