Classification on SVHN
3.2Error RateCAMNet4
Evaluation results
| Method | Links | ||
|---|---|---|---|
| CAMNet4Params(M)=3.02019.07 | 3.2 | — | |
| MultiCNN3Params(M)=1.52019.07 | 3.23 | — | |
| CAMNet3Params(M)=2.02019.07 | 3.28 | — | |
| tinyCAMNet3Params(M)=0.472019.07 | 3.38 | — | |
| SENetParams(M)=0.52019.07 | 3.42 | — | |
| Cr-Stitch2Params(M)=12019.07 | 3.45 | — | |
| CAMNet2Params(M)=1.22019.07 | 3.54 | — | |
| ADA-Net+Backbone=Conv-Large, Combined with=VAT+Ent2019.05 | 3.54 | — | |
| VAT+Ent+SNTGBackbone=Conv-Large2019.05 | 3.83 | — | |
| BaseCNNParams(M)=0.52019.07 | 3.86 | — | |
| VAT+EntBackbone=Conv-Large2019.05 | 3.86 | — | |
| Mean TeacherBackbone=Conv-Large2019.05 | 3.95 | — | |
| MA-DNNBackbone=Conv-Large2019.05 | 4.21 | — | |
| Temporal ensemblingBackbone=Conv-Large2019.05 | 4.42 | — | |
| ADA-NetBackbone=Conv-Large2019.05 | 4.62 | — | |
| SaaSBackbone=Conv-Large2019.05 | 4.77 | — | |
| II ModelBackbone=Conv-Large2019.05 | 4.82 | — | |
| VATBackbone=Conv-Large2019.05 | 5.42 | — | |
| uSF+BN+BMFeedback=Uniform Sign-concordant, Batch Normalization=true, Batch Manhattan=true2015.10 | 9.67 | — | |
| brSF+BN+BMFeedback=Biased Random Sign-concordant, Batch Normalization=true, Batch Manhattan=true2015.10 | 9.95 | — | |
| frSF+BN+BMFeedback=Fixed Random Sign-concordant, Batch Normalization=true, Batch Manhattan=true2015.10 | 10.16 | — | |
| Gumbel-ST2026.05 | — | 67.06 | |
| Multilinear-CovJac2026.05 | — | 69.19 | |
| Multilinear-STE2026.05 | — | 67.01 | |
| Soft-Mix2026.05 | — | 68.37 |