Research map / Continual learning / Domain and prompt-based
General Incremental Learning With Domain-Aware Categorical Representations
Check your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's title only (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| Add the domain-aware method to DER | DER · iCIFAR-20 ND | average incremental accuracy (%): 74.87 -> 84.03 | Helped |
| Add the domain-aware method to DER | DER · iCIFAR-20 NCD | average incremental accuracy (%): 74.56 -> 82.11 | Helped |
| Add the domain-aware method to DER | DER · iDomainNet ND | average incremental accuracy (%): 46.76 -> 61.05 | Helped |
| Add the domain-aware method to DER | DER · iDigits ND | average incremental accuracy (%): 93.62 -> 97.07 | Helped |
| Add the domain-aware method to UCIR | UCIR · iDigits NCD | average incremental accuracy (%): 86.29 -> 90.42 | Helped |
| Add a mixture model to DER (ablation) | DER · iCIFAR-20 NCD | average accuracy (%): 70.64 -> 78.71 | Helped |
| Add expansion-and-reduction to the mixture model (ablation) | DER with mixture model · iCIFAR-20 ND | average accuracy (%): 80.39 -> 81.28 | Helped |
| Add component regularization loss (ablation) | DER with mixture model and expansion-reduction · iCIFAR-20 NCD | average accuracy (%): 80.26 -> 80.97 | Helped |
| Add bi-level balanced memory (ablation) | DER with mixture model, expansion-reduction, and regularization · iCIFAR-20 ND | average accuracy (%): 82.04 -> 84.11 | Helped |
| Increase the number of added components from m=1 to m=10 (ablation) | DER w/ ours · iCIFAR-20 ND | average accuracy (%): 86.34 -> 87.40 | Helped |
| Increase memory size from M=500 to M=3000 (ablation) | DER w/ ours · iCIFAR-20 NCD | average accuracy (%): 75.04 -> 82.80 | Helped |
| Vary mixture-reduction threshold δ from 0.5 to 0.75 (ablation) | DER w/ ours · iCIFAR-20 NCD | average accuracy (%): 80.70 -> 82.05 | Mixed |
Closest papers by meaning
- Long-Tail Class Incremental Learning via Independent Sub-prototype Construction (CVPR 2024 · similarity 0.89)
- Non-exemplar Online Class-Incremental Continual Learning via Dual-Prototype Self-Augment and Refinement (AAAI 2024 · similarity 0.89)
- Addressing Imbalanced Domain-Incremental Learning through Dual-Balance Collaborative Experts (ICML 2025 · similarity 0.89)
- Task Confusion and Catastrophic Forgetting in Class-Incremental Learning: A Mathematical Framework for Discriminative and Generative Modelings (NeurIPS 2024 · similarity 0.88)
- Adaptive Discovering and Merging for Incremental Novel Class Discovery (AAAI 2024 · similarity 0.88)
- Make Domain Shift a Catastrophic Forgetting Alleviator in Class-Incremental Learning (AAAI 2025 · similarity 0.88)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.