Research map / Bandits and online learning / Contextual bandits and Thompson sampling
Multinomial Logit Contextual Bandits: Provable Optimality and Practicality
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 episodic updates and targeted random exploration instead of updating the parameter every round | MNL contextual bandit · Runtime, N=100, K=5, d=5, T=1000 | runtime (sec): UCB-MNL: 6.62; DBL-MNL: 1.20 | Helped |
| Use episodic updates and targeted random exploration instead of updating the parameter every round | MNL contextual bandit · Runtime, N=100, K=5, d=5, T=5000 | runtime (sec): UCB-MNL: 74.28; DBL-MNL: 5.92 | Helped |
| Add online approximate parameter updates in place of repeated exact MLE computation (ablation) | UCB-MNL · MNL contextual bandits | expected regret | No clear effect |
| Use episodic estimation and targeted exploration to obtain the sharper dimension dependence | DBL-MNL · MNL contextual bandits with uniform item revenue | expected regret: \tilde{\Ocal}(d\sqrt{T}) -> \tilde{\Ocal}(\sqrt{dT}) | Helped |
| Use the Auer-framework to generate independent samples | MNL contextual bandit · MNL contextual bandits | expected regret: \tilde{\Ocal}(\sqrt{dT}) | Helped |
Closest papers by meaning
- Nearly Minimax Optimal Regret for Multinomial Logistic Bandit (NeurIPS 2024 · similarity 0.96)
- Tractable Multinomial Logit Contextual Bandits with Non-Linear Utilities (NeurIPS 2025 · similarity 0.95)
- Contextual Multinomial Logit Bandits with General Value Functions (NeurIPS 2024 · similarity 0.95)
- Improved Online Confidence Bounds for Multinomial Logistic Bandits (ICML 2025 · similarity 0.90)
- Combinatorial Reinforcement Learning with Preference Feedback (ICML 2025 · similarity 0.90)
- Local Anti-Concentration Class: Logarithmic Regret for Greedy Linear Contextual Bandit (NeurIPS 2024 · similarity 0.90)
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.