Classification on RotMNIST (test)
99.28Classification AccuracyPartial G-CNN
Evaluation results
| Method | Links | ||
|---|---|---|---|
| Partial G-CNNBase Group=SE(2), Number of Elements=8, Partial Equivariance=true, Augerino=false2021.10 | 99.28 | — | |
| G-CNNBASE GROUP=SE(2), No. ELEMS=16, PARTIAL EQUIV.=false2021.10 | 99.24 | — | |
| Partial G-CNNBASE GROUP=SE(2), No. ELEMS=8, PARTIAL EQUIV.=true2021.10 | 99.23 | — | |
| Partial G-CNNBASE GROUP=SE(2), No. ELEMS=16, PARTIAL EQUIV.=true2021.10 | 99.18 | — | |
| G-CNNBASE GROUP=SE(2), No. ELEMS=8, PARTIAL EQUIV.=false2021.10 | 99.17 | — | |
| Partial G-CNNBASE GROUP=SE(2), No. ELEMS=4, PARTIAL EQUIV.=true2021.10 | 99.13 | — | |
| G-CNNBASE GROUP=SE(2), No. ELEMS=4, PARTIAL EQUIV.=false2021.10 | 99.1 | — | |
| Partial G-CNNBase Group=SE(2), Number of Elements=2, Partial Equivariance=true, Augerino=false2021.10 | 98.94 | — | |
| Partial G-CNNBase Group=SE(2), Number of Elements=4, Partial Equivariance=true, Augerino=false2021.10 | 98.94 | — | |
| AugerinoBase Group=SE(2), Number of Elements=4, Partial Equivariance=false, Augerino=true2021.10 | 98.78 | — | |
| AugerinoBase Group=SE(2), Number of Elements=8, Partial Equivariance=false, Augerino=true2021.10 | 98.77 | — | |
| AugerinoBase Group=SE(2), Number of Elements=2, Partial Equivariance=false, Augerino=true2021.10 | 98.72 | — | |
| G-CNNBase Group=SE(2), Number of Elements=2, Partial Equivariance=false, Augerino=false2021.10 | 98.7 | — | |
| Partial G-CNNBASE GROUP=E(2), No. ELEMS=16, PARTIAL EQUIV.=true2021.10 | 98.58 | — | |
| G-CNNBase Group=SE(2), Number of Elements=8, Partial Equivariance=false, Augerino=false2021.10 | 98.54 | — | |
| G-CNNBase Group=SE(2), Number of Elements=4, Partial Equivariance=false, Augerino=false2021.10 | 98.43 | — | |
| G-CNNBASE GROUP=E(2), No. ELEMS=16, PARTIAL EQUIV.=false2021.10 | 98.35 | — | |
| G-CNNBASE GROUP=E(2), No. ELEMS=8, PARTIAL EQUIV.=false2021.10 | 98.14 | — | |
| Partial G-CNNBASE GROUP=E(2), No. ELEMS=8, PARTIAL EQUIV.=true2021.10 | 97.78 | — | |
| ResNetBASE GROUP=T(2), No. ELEMS=12021.10 | 97.23 | — | |
| CNNBase Group=T(2), Number of Elements=1, Architecture=13-layer CNN2021.10 | 96.9 | — | |
| CONV32+GIDInput size=32 x 322026.02 | 96.32 | — | |
| RIC-CNNInput size=32 x 322026.02 | 95.52 | — | |
| E(2)-CNNInput size=29 x 292026.02 | 94.37 | — | |
| GA-CNNInput size=28 x 282026.02 | 93.29 | — | |
| H-NetInput size=32 x 322026.02 | 92.44 | — | |
| B-CNNInput size=28 x 282026.02 | 88.29 | — | |
| ORNInput size=32 x 322026.02 | 80.01 | — | |
| RotEqNetInput size=28 x 282026.02 | 73.2 | — | |
| DEF-CNNInput size=32 x 322026.02 | 46.97 | — | |
| CNNInput size=32 x 322026.02 | 45.42 | — | |
| G-CNNInput size=28 x 282026.02 | 44.81 | — | |
| CIDATemporal DG Category=Discrete temporal domain generalization2024.05 | — | 8.3 | |
| DeepODETemporal DG Category=Continuous temporal domain generalization2024.05 | — | 48.6 | |
| DRAINTemporal DG Category=Discrete temporal domain generalization2024.05 | — | 59.1 | |
| DRAIN-ΔtTemporal DG Category=Discrete temporal domain generalization2024.05 | — | 57.2 | |
| IncFinetuneTemporal DG Category=Practical baseline2024.05 | — | 57.1 | |
| IRMTemporal DG Category=Practical baseline2024.05 | — | 8.6 | |
| KoodosTemporal DG Category=Proposed method2024.05 | — | 4.6 | |
| LastDomainTemporal DG Category=Practical baseline2024.05 | — | 74.2 | |
| OfflineTemporal DG Category=Practical baseline2024.05 | — | 6.6 | |
| TKNetsTemporal DG Category=Discrete temporal domain generalization2024.05 | — | 37.7 | |
| V-RExTemporal DG Category=Practical baseline2024.05 | — | 8.6 |