Research map / Label-efficient and robust learning / Zero-/few-shot and meta-learning
Multidimensional Belief Quantification for Label-Efficient Meta-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 |
|---|---|---|---|
| Use belief-oriented task selection instead of no task selection (ablation) | Units-ML · Omniglot, 20-way 1-shot | accuracy (%): 91.96$\pm$0.48 -> 93.17$\pm$0.25 | Helped |
| Use belief-oriented task selection instead of no task selection (ablation) | Units-ML · Omniglot, 20-way 5-shot | accuracy (%): 97.38$\pm$0.07 -> 97.72$\pm$0.18 | Helped |
| Use belief-oriented task selection instead of no task selection (ablation) | Units-ML · mini-ImageNet, 5-way 1-shot | accuracy (%): 51.38$\pm$0.33 -> 50.86$\pm$0.67 | Hurt |
| Use belief-oriented task selection instead of no task selection (ablation) | Units-ML · mini-ImageNet, 5-way 5-shot | accuracy (%): 66.75$\pm$0.42 -> 68.16$\pm$0.72 | Helped |
| Use belief-oriented task selection instead of no task selection (ablation) | Units-ML · CifarFS, 5-way 1-shot | accuracy (%): 59.80$\pm$0.31 -> 59.84$\pm$0.11 | No clear effect |
| Use belief-oriented task selection instead of no task selection (ablation) | Units-ML · CifarFS, 5-way 5-shot | accuracy (%): 76.15$\pm$0.35 -> 76.69$\pm$0.44 | Helped |
| Select predictions with vacuity below 0.2 | Units-ML · CifarFS, 5-way 5-shot | accuracy (%): around 85% | Helped |
| Select more informative query sets for labeling (ablation) | Units-ML · Omniglot, 5-way 1-shot, 10,000 labeled tasks | accuracy (%): 96.98 -> 98.23 | Helped |
| Select more informative query sets for labeling (ablation) | Units-ML · Omniglot, 5-way 5-shot, 10,000 labeled tasks | accuracy (%): 98.46 -> 99.16 | Helped |
| Select more informative query sets for labeling (ablation) | Units-ML · Omniglot, 20-way 1-shot, 10,000 labeled tasks | accuracy (%): 83.56 -> 83.45 | No clear effect |
| Use multi-query task selection under a limited labeling budget | Units-ML · Omniglot, 20-way 5-shot | accuracy (%) | Helped |
Closest papers by meaning
- Probabilistic Model-Agnostic Meta-Learning (NeurIPS 2018 · similarity 0.88)
- Meta-Learning without Memorization (ICLR 2020 · similarity 0.88)
- Task Agnostic Meta-Learning for Few-Shot Learning (CVPR 2019 · similarity 0.87)
- Post-hoc Uncertainty Learning Using a Dirichlet Meta-Model (AAAI 2023 · similarity 0.87)
- The Role of Global Labels in Few-Shot Classification and How to Infer Them (NeurIPS 2021 · similarity 0.87)
- Making Large Vision Language Models to Be Good Few-Shot Learners (AAAI 2025 · 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.