Research map / Bandits and online learning / Contextual bandits and Thompson sampling
Bounded Off-Policy Evaluation with Missing Data for Course Recommendation and Curriculum Design
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 |
|---|---|---|---|
| Optimize entrance requirements in curriculum design (T = 6) (ablation) | Algorithm 1 · Curriculum design | expected CGPA: 3.35 -> 3.71 | Helped |
| Choose longer term end date T = 7 instead of T = 6 (ablation) | Algorithm 1 · Curriculum design | expected CGPA: 3.71 -> 3.30 | Hurt |
| Small lambda (e.g. 0) in policy optimization (7) (ablation) | Off-policy course policy · Course scheduling | value and variance | Mixed |
| Large lambda in policy optimization (7) (ablation) | Off-policy course policy · Course scheduling | policy value | Hurt |
| Set lambda = 1 (ablation) | Off-policy course policy · Course scheduling | value vs variance | Helped |
| Increase probability of observing missing context-action pair (ablation) | Regression estimator with missing data · Course schedule evaluation | bias | Hurt |
| Increase probability of observing missing context-action pair (ablation) | Regression estimator with missing data · Course schedule evaluation | variance | Hurt |
| Small p(a, x) below 10-3 for missing pairs (ablation) | Regression estimator with missing data · Course schedule evaluation | variance | No clear effect |
Closest papers by meaning
- Bandit Guided Submodular Curriculum for Adaptive Subset Selection (NeurIPS 2025 · similarity 0.83)
- Personalized Exercise Recommendation with Semantically-Grounded Knowledge Tracing (NeurIPS 2025 · similarity 0.81)
- Multi-type MOOCs Recommendation: Leveraging Deep Multi-Relational Representation and Hierarchical Reasoning (AAAI 2025 · similarity 0.81)
- Cross-Domain Off-Policy Evaluation and Learning for Contextual Bandits (ICLR 2025 · similarity 0.81)
- GenAL: Generative Agent for Adaptive Learning (AAAI 2025 · similarity 0.81)
- Simple Projection-Free Algorithm for Contextual Recommendation with Logarithmic Regret and Robustness (NeurIPS 2026 · similarity 0.81)
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.