Incremental Learning 벤치마크
Incremental Learning on ImageNet-100 - 50 classes + 10 steps of 5 classes
Average Incremental Accuracy
- 2020-04-28 — PODNet: Average Incremental Accuracy 73.14
- 2021-02-18 — CCIL-SD: Average Incremental Accuracy 76.77
- 2021-03-31 — DER: Average Incremental Accuracy 77.73
- 2023-01-14 — RMM (ResNet-18): Average Incremental Accuracy 78.47
| Rank | Model | Average Incremental Accuracy | Paper | Code | Year |
|---|---|---|---|---|---|
| 1 | RMM (ResNet-18) | 78.47 | RMM: Reinforced Memory Management for Class-Incremental Learning | g-u-n/pycil · aimagelab/mammoth · yaoyaoliu/rmm · +1 | 2023 |
| 2 | DER | 77.73 | DER: Dynamically Expandable Representation for Class Incremental Learning | g-u-n/pycil · Rhyssiyan/DER-ClassIL.pytorch | 2021 |
| 3 | FOSTER | 77.54 | FOSTER: Feature Boosting and Compression for Class-Incremental Learning | g-u-n/pycil · G-U-N/ECCV22-FOSTER | 2022 |
| 4 | CCIL-SD | 76.77 | Essentials for Class Incremental Learning | sud0301/essentials_for_CIL | 2021 |
| 5 | PODNet | 73.14 | PODNet: Pooled Outputs Distillation for Small-Tasks Incremental Learning | g-u-n/pycil · arthurdouillard/incremental_learning.pytorch | 2020 |