Research map / Label-efficient and robust learning / Domain adaptation and generalization
Active Learning with Cross-Class Similarity Transfer
Read the paperCheck 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 cross-class similarity transfer plus transfer-aware sample selection over plain AL baselines | active learning · CIFAR10, AwA, aPY | accuracy (%) at 10th iteration: +5.62%, +5.33%, +4.27% over best baseline | Helped |
| use cross-class transfer to cut labeled samples needed for 90% accuracy | cross-class T-AL · CIFAR10, AwA, aPY | labeled samples: 48 -> 24; 180 -> 70; 168 -> 84 | Helped |
| add class-graph propagation to sample-graph only (lambda 0 -> 0.5) (ablation) | similarity propagation · CIFAR10 | accuracy | Helped |
| add sample-graph propagation to class-graph only (lambda 1 -> 0.5) (ablation) | similarity propagation · CIFAR10 | accuracy | Helped |
| increase transferred samples per class Q from small to moderate (ablation) | cross-class T-AL · AwA | accuracy | Helped |
| increase transferred samples per class Q to 400 (ablation) | cross-class T-AL · AwA | accuracy | Hurt |
| add diversity term (eta) and source-neighbor term (tau) to uncertainty sampling (ablation) | sample selection · aPY | accuracy | Helped |
Closest papers by meaning
- Active Learning with Cross-Class Knowledge Transfer (AAAI 2016 · similarity 0.93)
- Composite Active Learning: Towards Multi-Domain Active Learning with Theoretical Guarantees (AAAI 2024 · similarity 0.87)
- Towards Cost-Effective Learning: A Synergy of Semi-Supervised and Active Learning (CVPR 2025 · similarity 0.86)
- A Cross-Domain Benchmark for Active Learning (NeurIPS 2024 · similarity 0.86)
- Instance-wise Supervision-level Optimization in Active Learning (CVPR 2025 · similarity 0.85)
- Category-Aware Active Domain Adaptation (ICML 2024 · similarity 0.85)
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.