Research map / Bandits and online learning / Contextual bandits and Thompson sampling
Statistical Inference with M-Estimators on Adaptively Collected 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 |
|---|---|---|---|
| Add square-root importance weights to least squares | Least-squares confidence regions · Continuous rewards, contextual bandit simulation | 90% confidence-region coverage | Helped |
| Remove square-root importance weighting (ablation) | AW-LS · Continuous rewards, contextual bandit simulation | 90% confidence-region coverage | Hurt |
| Add square-root importance weights to maximum likelihood | Maximum-likelihood confidence regions · Binary rewards, logistic regression simulation | 90% confidence-region coverage | Helped |
| Remove square-root importance weighting (ablation) | AW-MLE · Binary rewards, logistic regression simulation | 90% confidence-region coverage | Hurt |
| Use projected rather than non-projected confidence regions | AW-LS and AW-MLE advantage-parameter regions · Advantage parameters in continuous and binary reward simulations | Confidence-region volume | Mixed |
| Use AW-LS instead of the self-normalized martingale bound | Confidence regions for model parameters · Linear continuous-reward simulation | Confidence-region volume | Helped |
| Use AW-LS instead of the W-decorrelated estimator | Confidence regions for model parameters · Linear continuous-reward simulation | Confidence-region volume | Helped |
Closest papers by meaning
- Post-Contextual-Bandit Inference (NeurIPS 2021 · similarity 0.90)
- Simulation-Based Inference for Adaptive Experiments (NeurIPS 2025 · similarity 0.85)
- Robustly Improving Bandit Algorithms with Confounded and Selection Biased Offline Data: A Causal Approach (AAAI 2024 · similarity 0.84)
- Optimistic Algorithms for Adaptive Estimation of the Average Treatment Effect (ICML 2025 · similarity 0.84)
- Mixed-Effects Contextual Bandits (AAAI 2024 · similarity 0.83)
- Noise-Adaptive Confidence Sets for Linear Bandits and Application to Bayesian Optimization (ICML 2024 · similarity 0.83)
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.