Research map / Label-efficient and robust learning / Recognition and long-tail
Algorithm Selection for Deep Active Learning with Imbalanced Datasets
Jifan Zhang, Shuai Shao, saurabh verma, Robert D Nowak
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 |
|---|---|---|---|
| adaptive Thompson-sampling selection over candidate algorithms instead of a single best algorithm | TAILOR · CelebA multi-label classification | mean average precision | Improved |
| use TAILOR instead of the best single baseline algorithm | deep active learning · ten imbalanced datasets | accuracy | No clear effect |
| use TAILOR instead of the best baseline | deep active learning · ten imbalanced datasets | class diversity (labels in rarest class) | Improved |
| use TAILOR instead of confidence sampling | deep active learning · Caltech256 | balanced accuracy | Hurt |
| ALBL sampling (training-accuracy rewards) instead of Random Meta | meta algorithm selection · all datasets | accuracy | No clear effect |
| TAILOR instead of existing meta algorithms (ALBL, Random Meta) | meta algorithm selection · all ten datasets | accuracy and class diversity | Improved |
| TAILOR instead of best baseline for active search | multi-label search · CelebA, COCO, VOC, Car | total positive labels | Improved |
| run TAILOR rather than the slowest candidate algorithm | TAILOR · all experiments | running time: $20\%$ | Improved |
| Modified Submodular without balanced holdout set (ablation) | SIMILAR · multi-class classification | accuracy | Hurt |
Closest papers by meaning
- Towards Robust and Reproducible Active Learning Using Neural Networks (CVPR 2022 · similarity 0.86)
- Improved Algorithm for Deep Active Learning under Imbalance via Optimal Separation (ICML 2025 · similarity 0.86)
- Active Learning with Neural Networks: Insights from Nonparametric Statistics (NeurIPS 2022 · similarity 0.86)
- Active Learning Guided by Efficient Surrogate Learners (AAAI 2024 · similarity 0.85)
- One-shot Active Learning Based on Lewis Weight Sampling for Multiple Deep Models (ICLR 2024 · similarity 0.85)
- Instance-wise Supervision-level Optimization in Active Learning (CVPR 2025 · similarity 0.84)
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.
