Deep Clustering on CIFAR-100
0.921SCDCAM
Evaluation results
| Method | Links | ||||||
|---|---|---|---|---|---|---|---|
| DCAMselection_criterion=Best RRL while SC is within 10% of the best SC of the method2026.01 | 0.921 | 27.5 | — | — | — | — | |
| DCAMselection_criterion=Per-method best SC, architecture=Best across architectures2026.01 | 0.598 | -1.4 | — | — | — | — | |
| DCECselection_criterion=Best RRL while SC is within 10% of the best SC of the method2026.01 | 0.572 | 18.6 | — | — | — | — | |
| DCECselection_criterion=Per-method best SC, architecture=Best across architectures2026.01 | 0.47 | 7.5 | — | — | — | — | |
| NNMselection_criterion=Per-method best SC, architecture=Best across architectures2026.01 | 0.358 | — | — | — | — | — | |
| NNMselection_criterion=Best RRL while SC is within 10% of the best SC of the method2026.01 | 0.358 | — | — | — | — | — | |
| DCAMEncoder=CAE2026.01 | 0.354 | 267 | 0.219 | 0.0257 | 6.33 | — | |
| EDCselection_criterion=Best RRL while SC is within 10% of the best SC of the method2026.01 | 0.337 | 33.3 | — | — | — | — | |
| SCANselection_criterion=Per-method best SC, architecture=Best across architectures2026.01 | 0.321 | — | — | — | — | — | |
| SCANselection_criterion=Best RRL while SC is within 10% of the best SC of the method2026.01 | 0.321 | — | — | — | — | — | |
| EDCselection_criterion=Per-method best SC, architecture=Best across architectures2026.01 | 0.311 | 10 | — | — | — | — | |
| DCAMEncoder=EAE2026.01 | 0.2 | 160 | 0.225 | 0.025 | 6.33 | — | |
| Agglo.selection_criterion=Per-method best SC, architecture=Best across architectures2026.01 | 0.149 | — | — | — | — | — | |
| Agglo.selection_criterion=Best RRL while SC is within 10% of the best SC of the method2026.01 | 0.149 | — | — | — | — | — | |
| DCAMEncoder=RAE2026.01 | 0.13 | 465 | 0.239 | 0.0226 | 6.36 | — | |
| ClAMselection_criterion=Per-method best SC, architecture=Best across architectures2026.01 | 0.053 | — | — | — | — | — | |
| ClAMselection_criterion=Best RRL while SC is within 10% of the best SC of the method2026.01 | 0.053 | — | — | — | — | — | |
| DEKMselection_criterion=Best RRL while SC is within 10% of the best SC of the method2026.01 | 0.047 | 870 | — | — | — | — | |
| k-meansselection_criterion=Per-method best SC, architecture=Best across architectures2026.01 | 0.015 | — | — | — | — | — | |
| k-meansselection_criterion=Best RRL while SC is within 10% of the best SC of the method2026.01 | 0.015 | — | — | — | — | — | |
| DEKMselection_criterion=Per-method best SC, architecture=Best across architectures2026.01 | -0.007 | -34.3 | — | — | — | — |