Research map / Probabilistic and scientific modeling / Causal discovery
Degree of Interference: A General Framework For Causal Inference Under Interference
Yuki Ohnishi, Bikram Karmakar, Arman Sabbaghi
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 DoI Bayesian estimator instead of HT estimator | Bernoulli-trial E-ATE estimation · Scenario 1 (well specified), ER(p=0.01), N=300 | MSE (E-ATE): 0.0513 -> 0.0136 | Improved |
| use DoI Bayesian estimator instead of HT estimator | Bernoulli-trial E-ATE estimation · Scenario 1 (well specified), ER(p=0.01), N=1000 | MSE (E-ATE): 0.0156 -> 0.0033 | Improved |
| use DoI estimator instead of HT with simplified location function mu_k | E-ATE estimation · Scenario 2, ER(p=0.01), N=300 | MSE (E-ATE): 0.0508 -> 0.0134 | Improved |
| use DoI estimator instead of HT under multiplicative exp misspecification | E-ATE estimation · Scenario 3, ER(p=0.01), N=300 | MSE (E-ATE): 0.0678 -> 0.0132 | Improved |
| use DoI estimator instead of HT under exp misspecification | E-ATE estimation · Scenario 3, ER(p=0.01), N=1000 | MSE (E-ATE): 0.0168 -> 0.0033 | Improved |
| use DoI estimator instead of HT under exp treatment-spillover interaction | E-ATE estimation · Scenario 4, ER(p=0.01), N=1000 | MSE (E-ATE): 0.0152 -> 0.0034 | Improved |
| use DoI estimator instead of HT under cos misspecification | E-ATE estimation · Scenario 5, ER(p=0.01), N=1000 | MSE (E-ATE): 0.0152 -> 0.0032 | Improved |
| use DoI estimator instead of HT with unmeasured covariate U | E-ATE estimation · Scenario 6, ER(p=0.01), N=300 | MSE (E-ATE): 0.0775 -> 0.0279 | Improved |
| use DoI estimator instead of HT with unmeasured covariate U | E-ATE estimation · Scenario 6, ER(p=0.01), N=1000 | MSE (E-ATE): 0.0213 -> 0.0076 | Improved |
| use DoI estimator instead of HT on Barabasi-Albert graph | E-ATE estimation · Scenario 1, BA(n0=10,k=3), N=1000 | MSE (E-ATE): 0.0154 -> 0.0033 | Improved |
| increase sample size from 300 to 1000 for DoI | DoI estimator · Scenario 1, ER(p=0.01) | MSE (E-ATE): 0.0136 -> 0.0033 | Improved |
| add unmeasured covariate U to the data-generating process | DoI estimator · Scenario 1 vs Scenario 6, ER(p=0.01), N=1000 | MSE (E-ATE): 0.0033 -> 0.0076 | Hurt |
Closest papers by meaning
- Higher-Order Causal Message Passing for Experimentation with Complex Interference (NeurIPS 2024 · similarity 0.88)
- Independent-Set Design of Experiments for Estimating Treatment and Spillover Effects under Network Interference (ICLR 2024 · similarity 0.87)
- Online Experimental Design With Estimation-Regret Trade-off Under Network Interference (NeurIPS 2025 · similarity 0.87)
- Detecting and Measuring Confounding Using Causal Mechanism Shifts (NeurIPS 2024 · similarity 0.87)
- Data Fusion for Partial Identification of Causal Effects (NeurIPS 2025 · similarity 0.87)
- Learning Mixtures of Unknown Causal Interventions (NeurIPS 2024 · similarity 0.86)
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.
