Research map / Probabilistic and scientific modeling / Treatment effects
A causal fused lasso for interpretable heterogeneous treatment effects estimation
Oscar Hernan Madrid Padilla, Yanzhen Chen, Carlos Misael Madrid Padilla, Gabriel Ruiz
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 CFL1 (prognostic-score fused lasso) instead of GRF | causal fused lasso · Scenario 4 (n=1600, d=10) | MSE of treatment effect: 1.291 -> 0.277 | Improved |
| use CFL1 instead of GRF at high dimension | causal fused lasso · Scenario 2 (n=1600, d=10) | MSE of treatment effect: 0.350 -> 0.319 | Improved |
| use CFL1 instead of GRF at low dimension | causal fused lasso · Scenario 2 (n=1600, d=2) | MSE of treatment effect: 0.063 -> 0.108 | Hurt |
| use CFL2 (propensity-score ordering) instead of CFL1 | causal fused lasso · Scenario 3 (n=1600, d=10) | MSE of treatment effect: 0.293 -> 0.109 | Improved |
| use CFL1 instead of GRF when true effect is zero | causal fused lasso · Scenario 1 (n=1600, d=10) | MSE of treatment effect: 0.006 -> 0.003 | No clear effect |
| use CFL1 instead of ACW on large sample with sharp discontinuities | causal fused lasso · Scenario 6 (n=4000, d=10) | MSE of treatment effect: 0.131 -> 0.074 | Improved |
| use CFL1 instead of ACW on high-noise setting | causal fused lasso · Scenario 5 (n=4000, d=10) | MSE of treatment effect: 0.462 -> 0.078 | Improved |
| use CFL1 instead of ACW on JTPA semi-synthetic data | causal fused lasso · Scenario 7 (JTPA) | MSE of treatment effect: 0.015 -> 0.042 | Hurt |
| use CFL1 instead of ACW on Project STAR null effect data | causal fused lasso · Scenario 8 (Project STAR) | MSE of treatment effect: 0.501 -> 0.308 | Improved |
| use CFL2 instead of BART on nonlinear synthetic data | causal fused lasso · Scenario 9 | MSE of treatment effect: 0.174 -> 0.189 | Hurt |
| use CFL2 instead of ACW on Scenario 5-type data | causal fused lasso · n=4000, d=10 (appendix) | MSE of treatment effect: 0.471 -> 0.081 | Improved |
| use CFL1 instead of AIPW at high dimension | causal fused lasso · Scenario 4 (n=1600, d=10) | MSE of treatment effect: 3.134 -> 0.277 | Improved |
Closest papers by meaning
- Hierarchical and Density-based Causal Clustering (NeurIPS 2024 · similarity 0.88)
- Learning Joint Interventional Effects from Single-Variable Interventions in Additive Models (ICML 2025 · similarity 0.87)
- Doubly robust identification of treatment effects from multiple environments (ICLR 2025 · similarity 0.87)
- Data Fusion for Partial Identification of Causal Effects (NeurIPS 2025 · similarity 0.87)
- An Analysis of Causal Effect Estimation using Outcome Invariant Data Augmentation (NeurIPS 2025 · similarity 0.86)
- On Efficient Estimation of Distributional Treatment Effects under Covariate-Adaptive Randomization (ICML 2025 · 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.
