Research map / Bandits and online learning / Contextual bandits and Thompson sampling
Active Learning for Accurate Estimation of Linear Models
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 |
|---|---|---|---|
| Include the contextual trace term in the allocation score | Trace-UCB versus Var-UCB · Synthetic data, increasing dimension | median maximum prediction error | Helped |
| Include the contextual trace term in the allocation score | Trace-UCB versus Var-UCB · Synthetic data, n from 115 to 360 | expected maximum error: 25% to 50% higher errors | Helped |
| Include the contextual trace term in the allocation score | Trace-UCB versus Var-UCB · Synthetic data, high-probability loss, equal variances | median maximum prediction error: 25% | Helped |
| Use Trace-UCB instead of uniform allocation | Jester · Jester | prediction error: 45% to almost 20% | Helped |
| Use Trace-UCB instead of Var-UCB | Jester · Jester | prediction error: almost 30% to roughly 5% | Helped |
| Use Trace-UCB instead of uniform allocation | Trace-UCB · MovieLens | prediction error | Helped |
| Use Trace-UCB instead of Var-UCB | Trace-UCB · MovieLens | prediction error | No clear effect |
| Remove the contextual trace term from the allocation score (ablation) | Trace-UCB · Synthetic data, especially when dimension or instance count is large relative to the budget | prediction error | Hurt |
Closest papers by meaning
- Transductive Active Learning: Theory and Applications (NeurIPS 2024 · similarity 0.86)
- Active Context Selection Improves Simple Regret in Contextual Bandits (NeurIPS 2026 · similarity 0.86)
- Robust Sampling for Active Statistical Inference (NeurIPS 2025 · similarity 0.86)
- Uncertainty Herding: One Active Learning Method for All Label Budgets (ICLR 2025 · similarity 0.85)
- Noise-Adaptive Confidence Sets for Linear Bandits and Application to Bayesian Optimization (ICML 2024 · similarity 0.85)
- Improved Active Learning via Dependent Leverage Score Sampling (ICLR 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.