Research map / Probabilistic and scientific modeling / Treatment effects
A/B Testing for Recommender Systems in a Two-sided Marketplace
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 |
|---|---|---|---|
| Replace number of shared edges with normalized shared edges for candidate generation | Edge recommendation · Edge recommendation, viewee side | Weekly Active Unique users: +0.51% | Helped |
| Replace number of shared edges with normalized shared edges for candidate generation | Edge recommendation · Edge recommendation, viewee side | Sessions: +0.57% | Helped |
| Boost viewees likely to visit after receiving an edge formation request | Edge recommendation ranking model · Edge recommendation, viewee side | Weekly Active Unique users: +0.13% | Helped |
| Boost viewees likely to visit after receiving an edge formation request | Edge recommendation ranking model · Edge recommendation, viewee side | Sessions: +0.11% | Helped |
| Use UniCoRn rank-based reranking instead of modified OASIS score-based reranking | Simulated recommendation environment · Treatment effect estimation simulation | ATE estimation error | Helped |
| Use UniCoRn(1) or UniCoRn(0.2) instead of HaThucEtAl | Simulated recommendation environment · Treatment effect estimation simulation | ATE estimation error | Helped |
| Choose UniCoRn(0) instead of costlier UniCoRn variants at treatment proportion 0.5 (ablation) | UniCoRn · Treatment effect estimation simulation, TP 0.5 | ATE estimation error | No clear effect |
| Increase α from 0 to 1 (ablation) | UniCoRn · Simulated ranking design | RMSE | Helped |
| Increase α from 0 to 1 (ablation) | UniCoRn · Simulated ranking design | MAE | Mixed |
Closest papers by meaning
- Interference Among First-Price Pacing Equilibria: A Bias and Variance Analysis (ICLR 2025 · similarity 0.84)
- Recommender Ecosystems: A Mechanism Design Perspective on Holistic Modeling and Optimization (AAAI 2024 · similarity 0.83)
- Policy Design for Two-sided Platforms with Participation Dynamics (ICML 2025 · similarity 0.83)
- Addressing Hidden Confounding with Heterogeneous Observational Datasets for Recommendation (NeurIPS 2024 · similarity 0.83)
- Just Ramp-Up: Debiasing Regression-based Estimator for A/B Tests under Network Interference (NeurIPS 2026 · similarity 0.82)
- User-item fairness tradeoffs in recommendations (NeurIPS 2024 · similarity 0.82)
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.