Research map / Label-efficient and robust learning / Zero-/few-shot and meta-learning
Rethinking Class Relations: Absolute-Relative Supervised and Unsupervised Few-Shot Learning
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 Absolute-relative Learning | MetaOptNet · miniImageNet 5-way 1-shot | accuracy: 62.64±0.61 -> 65.21±0.58 | Helped |
| Add Absolute-relative Learning | SoSN · miniImageNet 5-way 1-shot and 5-shot | accuracy: 59.01±0.83 -> 61.36 ± 0.67; 75.49 ± 0.68 -> 78.95 ± 0.42 | Helped |
| Add Absolute-relative Learning | SoSN · CUB-200-2011 5-way 1-shot and 5-shot | accuracy: 46.72 -> 50.62; 60.34 -> 65.87 | Helped |
| Add Absolute-relative Learning | SoSN · Flower102 5-way 1-shot and 5-shot | accuracy: 71.90 -> 76.21; 84.87 -> 88.36 | Helped |
| Add unsupervised Absolute-relative Learning | U-SoSN · miniImageNet 5-way 1-shot and 5-shot | accuracy: 37.94 ± 0.87 -> 41.13 ± 0.84; 50.95 ± 0.81 -> 55.39 ± 0.79 | Helped |
| Add unsupervised Absolute-relative Learning | U-SoSN · Flower102 5-way 1-shot and 5-shot | accuracy: 61.26 -> 69.14; 75.98 -> 84.10 | Helped |
| Use attribute semantic labels for Relative Learning (ablation) | SoSN · miniImageNet 5-way | accuracy: 53.73 -> 55.56; 68.58 -> 70.97 | Helped |
| Use attribute prediction for Absolute Learning (ablation) | SoSN · miniImageNet 5-way | accuracy: 53.73 -> 55.31; 68.58 -> 71.03 | Helped |
| Use word2vec rather than attribute labels for Relative Learning (ablation) | SoSN · miniImageNet 5-way | accuracy: 54.31 vs 54.49; 69.64 vs 70.21 | Mixed |
| Combine class, attribute, and word2vec relative labels (ablation) | SoSN · miniImageNet 5-way | accuracy: 53.73 -> 55.49; 68.58 -> 70.86 | Helped |
| Use class-label Absolute Learning on fine-grained data (ablation) | SoSN · CUB-200-2011 5-way 1-shot | accuracy: 46.72 -> 46.88 | No clear effect |
| Use attribute Absolute Learning (ablation) | SoSN · CUB-200-2011 5-way 1-shot | accuracy: 46.72 -> 48.85 | Helped |
Closest papers by meaning
- Learning to Compare: Relation Network for Few-Shot Learning (CVPR 2018 · similarity 0.90)
- Meta-Learning for Semi-Supervised Few-Shot Classification (ICLR 2018 · similarity 0.89)
- Recovering the Missing Link: Predicting Class-Attribute Associations for Unsupervised Zero-Shot Learning (CVPR 2016 · similarity 0.88)
- Envisioning Class Entity Reasoning by Large Language Models for Few-shot Learning (AAAI 2025 · similarity 0.87)
- Class Semantic Attribute Perception Guided Zero-Shot Learning (AAAI 2025 · similarity 0.86)
- Simple Semantic-Aided Few-Shot Learning (CVPR 2024 · similarity 0.86)
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.