CLIQUE defines local feature importance as the average change in model error when a feature is replaced by values from its quantile grid, emphasizing locally dependent effects.
Panning for gold:‘model- x’knockoffs for high dimensional controlled vari- able selection
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
stat.ML 1years
2024 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
Conditional Local Importance by Quantile Expectations
CLIQUE defines local feature importance as the average change in model error when a feature is replaced by values from its quantile grid, emphasizing locally dependent effects.