Research map / Learning theory and algorithms / Prediction and decision losses
Learning Active Learning from Data
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 |
|---|---|---|---|
| Use a learned error-reduction regressor trained on synthetic 2D data instead of uncertainty sampling | LAL cold-start strategies · Striatum, MRI, Credit card | classification performance | Helped |
| Use iterative LAL rather than ALBE | LAL-iterative-2D · Checkerboard 2 x 2 | classification performance | Helped |
| Use iterative LAL rather than ALBE | LAL-iterative-2D · Banana | classification performance | No clear effect |
| Train LAL using application-specific warm-start data | LAL-independent-WS · Splice, Higgs | task performance | Helped |
| Use iterative LAL rather than independent LAL | LAL cold-start strategies · Striatum, MRI, Credit card | classification performance | Helped |
| Use LAL rather than uncertainty sampling | LAL · Checkerboard | seconds per AL iteration: 0.11 -> 0.54 | Hurt |
| Use LAL rather than ALBE | LAL · Checkerboard | seconds per AL iteration: 13.12 -> 0.54 | Helped |
| Use LAL rather than uncertainty sampling | LAL · MRI mini | seconds per AL iteration: 0.11 -> 0.55 | Hurt |
| Use LAL rather than ALBE | LAL · MRI mini | seconds per AL iteration: 64.52 -> 0.55 | Helped |
| Use LAL rather than uncertainty sampling | LAL · MRI | seconds per AL iteration: 0.12 -> 0.88 | Hurt |
| Use LAL rather than uncertainty sampling | LAL · Striatum mini | seconds per AL iteration: 0.11 -> 0.59 | Hurt |
| Use LAL rather than ALBE | LAL · Striatum mini | seconds per AL iteration: 75.64 -> 0.59 | Helped |
Closest papers by meaning
- Learning Algorithms for Active Learning (ICML 2017 · similarity 0.89)
- Towards Robust and Reproducible Active Learning Using Neural Networks (CVPR 2022 · similarity 0.89)
- Active Learning with Logged Data (ICML 2018 · similarity 0.88)
- Transductive Active Learning: Theory and Applications (NeurIPS 2024 · similarity 0.87)
- Provably Neural Active Learning Succeeds via Prioritizing Perplexing Samples (ICML 2024 · similarity 0.86)
- Active Classification with Few Queries under Misspecification (NeurIPS 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.