Class Incremental Learning On Cifar100
Metrics
10-stage average accuracy
Results
Performance results of various models on this benchmark
Model Name | 10-stage average accuracy | Paper Title | Repository |
|---|---|---|---|
| iCaRL | 63.24 | iCaRL: Incremental Classifier and Representation Learning | |
| A-GEM | 45.76 | Efficient Lifelong Learning with A-GEM | |
| EMR | 48.66 | On Tiny Episodic Memories in Continual Learning | |
| LUCIR | 56.53 | Learning a Unified Classifier Incrementally via Rebalancing | |
| ABD | 54.44 | Always Be Dreaming: A New Approach for Data-Free Class-Incremental Learning | |
| S&B | 68.18 | Split-and-Bridge: Adaptable Class Incremental Learning within a Single Neural Network | |
| SCR | 65.98 | Supervised Contrastive Learning |
0 of 7 row(s) selected.