PermuCATE applies conditional permutation importance to CATE estimation, claiming lower variance and higher statistical power than LOCO on simulated and health datasets.
However, the term E[( ε(X,A) π(X)(1−π(X)))2] will likely take extreme values and lead to inconsistencies between theRP O-risk and the oracle τ-MSE
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.LG 1years
2024 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
Measuring Variable Importance in Heterogeneous Treatment Effects with Confidence
PermuCATE applies conditional permutation importance to CATE estimation, claiming lower variance and higher statistical power than LOCO on simulated and health datasets.