Research map / Probabilistic and scientific modeling / Treatment effects
Cluster-Adaptive Network A/B Testing: From Randomization to Estimation
Yang Liu, Yifan Zhou, Ping Li, Feifang Hu
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 CAE instead of DIM estimator | CR design · Simulated network (m=200, r=1.8) | RMSE: 1.395 -> 0.698 | Improved |
| use CAE instead of DIM estimator | CLAR-Both design · Paluck school network | RMSE (x10^-2): 7.778 -> 7.082 | Improved |
| add covariate balancing (CLAR-Both instead of CR) before CAE (ablation) | CAE · Simulated network (m=200, r=1.8) | standard deviation: 0.698 -> 0.477 | Improved |
| add cluster-level covariates to CLAR (ablation) | CAE · Simulated network (m=200, r=1.8) | standard deviation: 0.698 -> 0.504 | Improved |
| add individual-level covariates only to CLAR (ablation) | CAE · Simulated network (m=200, r=1.8) | standard deviation: 0.698 -> 0.556 | Improved |
| add both covariate levels (CLAR-Both instead of CLAR-Ind) (ablation) | CAE · Paluck school network | RMSE (x10^-2): 7.872 -> 7.082 | Improved |
| use CLAR design with DIM estimator (no estimator adjustment) (ablation) | DIM · Simulated network | bias | No clear effect |
| increase interference r from 0.2 to 1.8 (ablation) | DIM estimator · Simulated network (m=200) | bias under CR: -0.127 -> -0.748 | Hurt |
| add CLAR-Both instead of CR (ablation) | CAE · Simulated network (m=50, r=0.2) | RMSE: 0.952 -> 0.663 | Improved |
| add CLAR-Both instead of CR (ablation) | CAE · Simulated network (m=100, r=1.8) | RMSE: 0.983 -> 0.683 | Improved |
| include individual-level covariates in CLAR (CLAR-Ind) versus CLAR-CL (ablation) | CAE · Paluck school network | RMSE (x10^-2): 7.594 -> 7.872 | Hurt |
Closest papers by meaning
- Just Ramp-Up: Debiasing Regression-based Estimator for A/B Tests under Network Interference (NeurIPS 2026 · similarity 0.89)
- Strategic A/B testing via Maximum Probability-driven Two-armed Bandit (ICML 2025 · similarity 0.84)
- Asymptotically Optimal and Computationally Efficient Average Treatment Effect Estimation in A/B testing (ICML 2024 · similarity 0.83)
- Network two-sample test for block models (NeurIPS 2025 · similarity 0.83)
- Independent-Set Design of Experiments for Estimating Treatment and Spillover Effects under Network Interference (ICLR 2024 · similarity 0.82)
- YEAST: Yet Another Sequential Test (NeurIPS 2025 · 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.
