CLAGA re-labels each training instance with out-of-sample CATE estimates from K-fold primary models, eliminating group-assignment-dependent predictions and reducing PEHE on several benchmarks.
From real-world patient data to individualized treatment effects using machine learning: current and future methods to address underlying challenges
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.LG 1years
2025 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
Consistent Labeling Across Group Assignments: Variance Reduction in Conditional Average Treatment Effect Estimation
CLAGA re-labels each training instance with out-of-sample CATE estimates from K-fold primary models, eliminating group-assignment-dependent predictions and reducing PEHE on several benchmarks.