Image Classification on M(CIFAR100-10, F-CelebA)
68.31AccuracyCAT
Evaluation results
| Method | Links | |||
|---|---|---|---|---|
| CATArchitecture=AlexNet-like, Number of random sequences=52021.12 | 68.31 | 60.99 | 75.64 | |
| ONEArchitecture=AlexNet-like, Number of random sequences=52021.12 | 67.64 | 68.92 | 66.36 | |
| HATArchitecture=AlexNet-like, Number of random sequences=52021.12 | 61.78 | 63.01 | 60.55 | |
| NCLArchitecture=AlexNet-like, Number of random sequences=52021.12 | 61.55 | 56.92 | 66.18 | |
| ONENetwork Architecture=2-layer fully connected network, Random Task Sequences=52021.12 | 58.61 | — | — | |
| HATNetwork Architecture=2-layer fully connected network, Random Task Sequences=52021.12 | 56.92 | — | — | |
| PRONetwork Architecture=2-layer fully connected network, Random Task Sequences=52021.12 | 55.99 | — | — | |
| CATNetwork Architecture=2-layer fully connected network, Random Task Sequences=52021.12 | 54.79 | — | — | |
| UCLNetwork Architecture=2-layer fully connected network, Random Task Sequences=52021.12 | 53.73 | — | — | |
| EWCNetwork Architecture=2-layer fully connected network, Random Task Sequences=52021.12 | 53.45 | — | — | |
| PathNetNetwork Architecture=2-layer fully connected network, Random Task Sequences=52021.12 | 52.44 | — | — | |
| NCLNetwork Architecture=2-layer fully connected network, Random Task Sequences=52021.12 | 51.24 | — | — | |
| HYPNetwork Architecture=2-layer fully connected network, Random Task Sequences=52021.12 | 46.67 | — | — | |
| RPSNetNetwork Architecture=2-layer fully connected network, Random Task Sequences=52021.12 | 40.56 | — | — | |
| HYP-RNetwork Architecture=2-layer fully connected network, Random Task Sequences=52021.12 | 20.96 | — | — |