iLOCO estimates the importance of pairwise and higher-order feature interactions in any predictive model and provides distribution-free confidence intervals via data splitting or minipatch ensembles.
Permutation importance: a corrected feature importance measure
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
stat.ML 1years
2025 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
iLOCO: Distribution-Free Inference for Feature Interactions
iLOCO estimates the importance of pairwise and higher-order feature interactions in any predictive model and provides distribution-free confidence intervals via data splitting or minipatch ensembles.