Research map / Probabilistic and scientific modeling / Causal discovery
Boosted Control Functions: Distribution Generalization and Invariance in Confounded Models
Nicola Gnecco, Jonas Peters, Sebastian Engelke, Niklas Pfister
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 BCF instead of least squares random forest under strong exogenous perturbation | Boosted Control Function · Experiment 1 simulated SIMDG, k in 1..10 | MSE | Improved |
| use BCF instead of the true structural function | Boosted Control Function · Experiment 1 simulated SIMDG | MSE | Improved |
| increase rank q of M0 (ablation) | BCF versus LS · Experiment 1 simulated SIMDG, large k | MSE | Improved |
| increase eigengap tau of M0 (ablation) | rank selection estimator of M0 · Experiment 2 subspace distance, q=5 | subspace distance | Improved |
| use BCF instead of LS and classical control function | Boosted Control Function · California housing geographical splits SW/(N+SE), (N+SW)/SE, N/S | test MSE | Improved |
| use BCF instead of Anchor Boosting | Boosted Control Function · California housing geographical splits | test MSE | Mixed |
| use BCF instead of GroupDRO | Boosted Control Function · California housing geographical splits | test MSE | No clear effect |
Closest papers by meaning
- Not all distributional shifts are equal: Fine-grained robust conformal inference (ICML 2024 · similarity 0.86)
- Benign Overfitting in Out-of-Distribution Generalization of Linear Models (ICLR 2025 · similarity 0.85)
- Optimal Nuisance Function Tuning for Estimating a Doubly Robust Functional under Proportional Asymptotics (NeurIPS 2025 · similarity 0.85)
- Impossibility of Distribution-Free Predictive Inference for Individual Treatment Effects (NeurIPS 2026 · similarity 0.85)
- Data-Driven Covariate Selection for Nonparametric and Cycle-Agnostic Causal Effect Estimation (NeurIPS 2026 · similarity 0.85)
- A Versatile Influence Function for Data Attribution with Non-Decomposable Loss (ICML 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.
