Research map / Bandits and online learning / Multi-armed bandits
Machine Learning for Online Algorithm Selection under Censored Feedback
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 Thompson sampling instead of LinUCB | Online algorithm selection under censored feedback · ASlib scenarios | rePAR10 | Helped |
| Use Thompson sampling rather than linear ε-greedy | Online algorithm selection under censored feedback · ASlib scenarios | rePAR10: around 6% | Helped |
| Add expected PAR10 loss minimization to Thompson sampling (ablation) | Thompson sampling · ASlib scenarios | rePAR10 | Helped |
| Add Buckley-James imputation to Thompson sampling (ablation) | Thompson sampling · ASlib scenarios | rePAR10 | Helped |
| Combine expected PAR10 loss minimization with Buckley-James imputation (ablation) | Thompson sampling · ASlib scenarios | rePAR10 | Hurt |
| Randomize the confidence widths (ablation) | LinUCB with bias-corrected confidence bounds · ASlib scenarios | rePAR10 | Helped |
| Use bias-corrected confidence bounds instead of uncorrected bounds (ablation) | LinUCB · ASlib scenarios | rePAR10 | Hurt |
| Add expected PAR10 loss minimization to LinUCB variants (ablation) | LinUCB · ASlib scenarios | rePAR10 | Mixed |
| Ignore censored observations during learning (ablation) | BlindUCB · ASlib scenarios | algorithm-selection performance | Helped |
| Use the combined revisited and Buckley-James variant instead of either component alone (ablation) | Thompson sampling · ASlib scenarios | rePAR10 | Hurt |
Closest papers by meaning
- Algorithmic Assurance: An Active Approach to Algorithmic Testing using Bayesian Optimisation (NeurIPS 2018 · similarity 0.86)
- The Stability of Online Algorithms in Performative Prediction (NeurIPS 2026 · similarity 0.85)
- Put CASH on Bandits: A Max K-Armed Problem for Automated Machine Learning (NeurIPS 2025 · similarity 0.85)
- Online Algorithms with Uncertainty-Quantified Predictions (ICML 2024 · similarity 0.85)
- Robustly Improving Bandit Algorithms with Confounded and Selection Biased Offline Data: A Causal Approach (AAAI 2024 · similarity 0.85)
- Bandit and Delayed Feedback in Online Structured Prediction (NeurIPS 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.