Research map / Learning theory and algorithms / Submodular and game algorithms
Learning User Preferences to Incentivize Exploration in the Sharing Economy
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 hemimetric structure to coordinate learning across item pairs | CoOL versus IOL · Airbnb user study | cumulative utility gain: almost 50% higher | Helped |
| Use weighted projections onto the hemimetric structure | CoOL versus IOL · Synthetic hemimetric data, random problem order, T=500 | regret: less than half | Helped |
| Use structural projections when problems arrive in batches of five | CoOL versus IOL · Synthetic hemimetric data, batches of problems | regret | Mixed |
| Use coordinated projections when the same single problem repeats | CoOL versus IOL · Synthetic hemimetric data, single problem repeated | regret | No clear effect |
| Replace frequency-weighted projections with unweighted projections (ablation) | CoOL · Synthetic hemimetric data, single problem repeated | regret | Hurt |
| Reduce the projection rate toward zero (ablation) | CoOL · Synthetic hemimetric data | regret: at α=0 | Hurt |
| Use projection rate α=0.1 instead of no projections (ablation) | CoOL · Synthetic hemimetric data | regret: almost half | Helped |
| Reduce projection accuracy by lowering β (ablation) | CoOL · Synthetic hemimetric data | regret: smaller than 0.65 | Hurt |
| Use approximate projections with β between 0.85 and 0.95 (ablation) | CoOL · Synthetic hemimetric data | runtime: less than 10% | Helped |
| Use approximate projections with β between 0.85 and 0.95 (ablation) | CoOL · Synthetic hemimetric data | regret: less than half | Helped |
Closest papers by meaning
- Decongestion by Representation: Learning to Improve Economic Welfare in Marketplaces (ICLR 2024 · similarity 0.86)
- Beyond Self-Interest: How Group Strategies Reshape Content Creation in Recommendation Platforms? (ICML 2025 · similarity 0.85)
- Capturing Individual Human Preferences with Reward Features (NeurIPS 2025 · similarity 0.85)
- Direct Alignment with Heterogeneous Preferences (NeurIPS 2025 · similarity 0.85)
- Learning from Streaming Data when Users Choose (ICML 2024 · similarity 0.85)
- The Burden of Interactive Alignment with Inconsistent Preferences (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.