Research map / Continual learning / Class-incremental
On the Approximation Risk of Few-Shot Class-Incremental Learning
Xuan Wang, Zhong Ji, Xiyao Liu, Yanwei Pang, Jungong Han
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (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 |
|---|---|---|---|
| discard the base classifier and train per-task classifiers independently instead of fine-tuning the base classifier (ablation) | FSCIL ViT-B/16 · CUB200 | final-session accuracy (%): 38.01 -> 70.97 | Improved |
| freeze the base classifier instead of fine-tuning it (ablation) | FSCIL ViT-B/16 · CUB200 | final-session accuracy (%): 38.01 -> 52.41 | Improved |
| add tanh after each normalization layer (Stabilize) (ablation) | ViT-B/16 with DS · CUB200 | DR (%): 18.47 -> 17.30 | Improved |
| add layer norm + tanh on shared representation (Constrain) (ablation) | ViT-B/16 with DS · CUB200 | DR (%): 18.47 -> 16.69 | Improved |
| add label smoothing (ablation) | ViT-B/16 with DS · CUB200 | DR (%): 18.47 -> 16.82 | Improved |
| combine Stabilize, Constrain and label smoothing (ablation) | ViT-B/16 with DS · CUB200 | final-session accuracy (%): 70.97 -> 75.22 | Improved |
| use ImageNet-1K pretrained ViT instead of CLIP-pretrained ViT | FSCIL ViT-B/16 · CUB200 | final-session accuracy (%): 75.22 -> 81.30 | Improved |
| increase base-class dataset size via data augmentation and Mixup | ResNet18 and ViT-S from scratch · mini-ImageNet | final accuracy | Improved |
| apply the guidelines on mini-ImageNet and CIFAR100 with ViTs | FSCIL ViT · mini-ImageNet, CIFAR100 | accuracy | Mixed |
Closest papers by meaning
- Few-Shot Incremental Learning With Continually Evolved Classifiers (CVPR 2021 · similarity 0.94)
- Few-Shot Class-Incremental Learning for Classification and Object Detection: A Survey (TPAMI 2025 · similarity 0.92)
- Adaptive Decision Boundary for Few-Shot Class-Incremental Learning (AAAI 2025 · similarity 0.92)
- Tripartite Weight-Space Ensemble for Few-Shot Class-Incremental Learning (CVPR 2025 · similarity 0.91)
- Pseudo Informative Episode Construction for Few-Shot Class-Incremental Learning (AAAI 2025 · similarity 0.88)
- Pre-trained Vision and Language Transformers Are Few-Shot Incremental Learners (CVPR 2024 · 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.
