Research map / Probabilistic and scientific modeling / Treatment effects
FAST: a Fused and Accurate Shrinkage Tree for Heterogeneous Treatment Effects Estimation
Jia Gu, Caizhi Tang, Han Yan, Qing Cui, Longfei Li, JUN ZHOU
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 |
|---|---|---|---|
| replace shrinkage tree split criterion with the proposed FAST split criterion (9) (ablation) | Shrinkage Tree · Simulation, n=100, beta=1.0 | RMSE: 1.90 -> 1.85 | Improved |
| add adaptive fusion split criterion (ST to FAST) (ablation) | Shrinkage Tree · Simulation, n=200, beta=1.0 | RMSE: 1.98 -> 1.83 | Improved |
| fuse observational data via shrinkage (ST) instead of trial-only honest tree | HT · Simulation, n=100, beta=1.0 | RMSE: 2.28 -> 1.90 | Improved |
| fuse observational data via shrinkage (ST) instead of trial-only honest tree | HT · Simulation, n=200, beta=1.0 | RMSE: 2.20 -> 1.98 | Improved |
| bag FAST into a random forest (rfFAST) | FAST · Simulation, n=100, beta=1.0 | RMSE: 1.85 -> 0.89 | Improved |
| bag FAST into a random forest (rfFAST) | FAST · Simulation, n=200, beta=1.0 | RMSE: 1.83 -> 0.84 | Improved |
| rfFAST instead of GRF trial-only forest | GRF · Simulation, n=100, beta=1.0 | RMSE: 1.19 -> 0.89 | Improved |
| rfFAST instead of GRF trial-only forest | GRF · Simulation, n=200, beta=1.0 | RMSE: 1.12 -> 0.84 | Improved |
| use KPS with OLS confounding function instead of GRF | GRF · Simulation, n=200, beta=1.0 | RMSE: 1.12 -> 0.97 | Improved |
| use KPS with OLS confounding function instead of GRF | GRF · Simulation, n=100, beta=1.0 | RMSE: 1.19 -> 1.29 | Hurt |
| use KPS with random-forest confounding function instead of GRF | GRF · Simulation, n=100, beta=1.0 | RMSE: 1.19 -> 1.65 | Hurt |
| simple fusion of trial and observational data at large confounding bias | Fit-based causal tree · Simulation, beta sweep 0.1-1.9 | RMSE | Hurt |
Later papers that compared with this method
1 later accepted paper reports FAST+ as a baseline in their results tables; 1 of them report a better number on most of the results they share with it.
- Transporting Quantiles to Codebook: Scalable Vector Quantization without Codebook Collapse (NeurIPS 2026 · better on 1 of 1 shared result)
Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.
Closest papers by meaning
- Estimating Distributional Treatment Effects in Randomized Experiments: Machine Learning for Variance Reduction (ICML 2024 · similarity 0.87)
- Uncertainty Quantification in Heterogeneous Treatment Effect Estimation with Gaussian-Process-Based Partially Linear Model (AAAI 2024 · similarity 0.86)
- On Efficient Estimation of Distributional Treatment Effects under Covariate-Adaptive Randomization (ICML 2025 · similarity 0.86)
- Estimating Heterogeneous Treatment Effects by Combining Weak Instruments and Observational Data (NeurIPS 2024 · similarity 0.86)
- Generalized Random Forests Using Fixed-Point Trees (ICML 2025 · similarity 0.86)
- Do Contemporary Causal Inference Models Capture Real-World Heterogeneity? Findings from a Large-Scale Benchmark (ICLR 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.
