AI RESEARCH

SHIFT: Robust Double Machine Learning for Average Dose-Response Functions under Heavy-Tailed Contamination

arXiv CS.LG

ArXi:2605.00176v1 Announce Type: cross Double-machine-learning pipelines for the Average Dose-Response Function rely on kernel-weighted local-linear smoothers, which inherit unbounded functional influence: a single outlier within a kernel window biases the curve across the entire window. We