SOURCE-LINKED INTELLIGENCE

Just add noise: Debiasing tree-based variable importance in mixed data

arXiv · AI, language, vision and robotics · article · Sep 12, 2026 · UTC

Variable importance scores from tree-based methods such as random forests favor continuous predictors over categorical ones. We present a theoretical analysis of this bias and propose a simple remedy: add a small amount of noise to each categorical predictor. The correction is demonstrated on a variety of simulated and real-world datasets and combined with integrated path stability selection to perform variable selection with false discovery control for mixed data.

Read original source ↗ Open in workspace

recordType

paper

region

Global