Research map / Learning theory and algorithms / Sample complexity
Distributed Statistical Inference under Heterogeneity
Jia Gu, Song Xi Chen
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 weighted distributed (WD) averaging instead of SaC | SaC · logistic regression simulation, p2=10, K=1000 | 95% CI coverage: .00 -> .37 | Improved |
| Add bias correction to WD (dWD) | WD · logistic regression simulation, p2=10, K=2000 | 95% CI coverage: .00 -> .93 | Improved |
| Add bias correction to SaC (dSaC) | SaC · logistic regression simulation, p2=10, K=2000 | 95% CI coverage: .00 -> .96 | Improved |
| Use dWD instead of dSaC for interval width | dSaC · logistic regression simulation, p2=10, K=2000 | 95% CI width: 2.45 -> 1.82 | Improved |
| Use WD weighting instead of SaC for interval width | SaC · logistic regression simulation, p2=10, K=2000 | 95% CI width: 2.45 -> 1.78 | Improved |
| Add bias correction to WD | WD · logistic regression simulation, p2=10, K=250 | 95% CI coverage: .92 -> .96 | No clear effect |
| Add weighting plus bias correction (dWD) vs SaC | SaC · logistic regression simulation, p2=10, K=2000 | CPU time: 23.42 -> 48.05 | Hurt |
| Add weighting (WD) vs SaC | SaC · logistic regression simulation, p2=10, K=2000 | CPU time: 23.42 -> 24.62 | No clear effect |
| Add bias correction to SaC (dSaC) | SaC · logistic regression simulation, p2=10, K=2000 | CPU time: 23.42 -> 30.43 | Hurt |
| Use WD instead of SaC | SaC · Airline on-time data, rain parameter | CI length: 19.1% shorter | Improved |
| Use WD instead of SaC | SaC · Airline on-time data, pressure and temperature | CI length: 2.2% and 2.9% shorter | Improved |
| Use SaC instead of SAVGM at large K | SAVGM · logistic regression simulation, p2=10 | RMSE | Improved |
Closest papers by meaning
- Least Squares Model Averaging for Distributed Data (JMLR 2023 · similarity 0.87)
- Communication-efficient Distributed Statistical Inference for Massive Data with Heterogeneous Auxiliary Information (JMLR 2026 · similarity 0.87)
- Doubly Distributed Supervised Learning and Inference with High-Dimensional Correlated Outcomes (JMLR 2020 · similarity 0.86)
- WONDER: Weighted One-shot Distributed Ridge Regression in High Dimensions (JMLR 2020 · similarity 0.86)
- One-Shot Weighted Ensemble Estimation for Federated Quantile Regression: Optimal Statistical Guarantees under Heterogeneous Structured Data (ICML 2026 · similarity 0.85)
- Selective Inference with Distributed Data (JMLR 2025 · similarity 0.85)
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.
