Research map / Bandits and online learning / Contextual bandits and Thompson sampling
Learning to Rank under Multinomial Logit Choice
James A. Grant, David S. Leslie
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (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 tighter Geometric concentration coefficients instead of weaker-inequality coefficients (Epoch-UCB vs Epoch-UCB-W) (ablation) | Epoch-UCB (known position biases) · MNL-LTR simulations (a)-(c) | cumulative regret | Improved |
| use modified UCB coefficient 0.5 scaling (Epoch-UCB* UPB) instead of unmodified Epoch-UCB UPB (ablation) | Epoch-UCB UPB (unknown position biases) · MNL-LTR simulations (a)-(c) | cumulative regret | Improved |
| learn position biases with the MNL-LTR structure (J+K-1 parameters) instead of treating J*K product parameters separately (MNL-bandit) (ablation) | Epoch-UCB* UPB · MNL-LTR problem (c), K=6, J=30 | cumulative regret: estimates J+K-1=35 parameters vs JK=180 | Improved |
| compare against TopRank, which assumes decreasing position biases, when position biases are not decreasing | TopRank · MNL-LTR problems (b) and (c) | cumulative regret | Hurt |
| compare TopRank to Epoch-UCB when position biases are decreasing | TopRank · MNL-LTR problem (a) | cumulative regret | No clear effect |
Closest papers by meaning
- Learning from Ranking Feedback: Improved Regret Bounds via Independence Preserving Rank Breaking (NeurIPS 2026 · similarity 0.88)
- Online Learning of Delayed Choices (NeurIPS 2024 · similarity 0.88)
- Adaptively Learning to Select-Rank in Online Platforms (ICML 2024 · similarity 0.88)
- Active preference learning for ordering items in- and out-of-sample (NeurIPS 2024 · similarity 0.86)
- On the Convergence of No-Regret Dynamics in Information Retrieval Games with Proportional Ranking Functions (ICLR 2025 · similarity 0.86)
- Nearly Minimax Optimal Regret for Multinomial Logistic Bandit (NeurIPS 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.
