Research map / Bandits and online learning / Contextual bandits and Thompson sampling
Off-Policy Risk Assessment in Contextual Bandits
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 the doubly robust CDF estimator instead of IS or WIS | OPRA CDF estimation · PageBlocks | sup-norm CDF error | Helped |
| Use the doubly robust CDF estimator instead of IS or WIS at small sample sizes | OPRA CDF estimation · OptDigits | sup-norm CDF error | Hurt |
| Use the doubly robust CDF estimator instead of IS or WIS as the model improves with more data | OPRA CDF estimation · OptDigits | sup-norm CDF error | Helped |
| Use the doubly robust estimator instead of IS or WIS | OPRA risk estimation · OptDigits | mean, CVaR, and variance estimation error | Mixed |
| Use the direct method instead of IS or DR | OPRA CDF estimation · PageBlocks | sup-norm CDF error | Mixed |
| Use the direct method instead of IS, WIS, or DR | OPRA risk estimation · OptDigits | risk estimate error | Hurt |
| Estimate one CDF and reuse it to evaluate multiple Lipschitz risk functionals | OPRA · Contextual bandit off-policy evaluation | risk-estimation error rate: O(1/√n) | Helped |
Closest papers by meaning
- Optimal and Adaptive Off-policy Evaluation in Contextual Bandits (ICML 2017 · similarity 0.92)
- Empirical Likelihood for Contextual Bandits (NeurIPS 2020 · similarity 0.91)
- Marginal Density Ratio for Off-Policy Evaluation in Contextual Bandits (NeurIPS 2023 · similarity 0.91)
- Conformal Off-Policy Prediction in Contextual Bandits (NeurIPS 2022 · similarity 0.91)
- Logarithmic Smoothing for Pessimistic Off-Policy Evaluation, Selection and Learning (NeurIPS 2024 · similarity 0.90)
- Log-Sum-Exponential Estimator for Off-Policy Evaluation and Learning (ICML 2025 · similarity 0.88)
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.