Class Incremental Learning on Tiny-ImageNet-200 T200-B100-S10 (test)
37.99Final AccuracyGIS (Ours)
Evaluation results
| Method | Links | |
|---|---|---|
| GIS (Ours)Backbone=ResNet-18, Batch size=64, Optimizer=Adam, Initial learning rate=0.001, Epochs per step=30, Memory buffer size=20 instances per class2023.04 | 37.99 | |
| IL2ABackbone=ResNet-18, Batch size=64, Optimizer=Adam, Initial learning rate=0.001, Epochs per step=30, Memory buffer size=20 instances per class2023.04 | 34.28 | |
| EEILBackbone=ResNet-18, Batch size=64, Optimizer=Adam, Initial learning rate=0.001, Epochs per step=30, Memory buffer size=20 instances per class2023.04 | 33.67 | |
| HFABackbone=ResNet-18, Batch size=64, Optimizer=Adam, Initial learning rate=0.001, Epochs per step=30, Memory buffer size=20 instances per class2023.04 | 33.65 | |
| iCaRLNMEBackbone=ResNet-18, Batch size=64, Optimizer=Adam, Initial learning rate=0.001, Epochs per step=30, Memory buffer size=20 instances per class2023.04 | 33.24 | |
| iCaRLCNNBackbone=ResNet-18, Batch size=64, Optimizer=Adam, Initial learning rate=0.001, Epochs per step=30, Memory buffer size=20 instances per class2023.04 | 20.71 |