SHIFT combines cross-fit DML with kernel-local Welsch loss optimized via Graduated Non-Convexity and a MAD-scaled defensive OLS refit to achieve robust average dose-response estimation under localized heavy-tailed contamination while recovering outlier masks.
If Hill ˆα <3or GPD ˆξ > 0: switch toQuantile-DMLfor the point estimate; retainSHIFT’s mask for the detection report
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
stat.ML 1years
2026 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
SHIFT: Robust Double Machine Learning for Average Dose-Response Functions under Heavy-Tailed Contamination
SHIFT combines cross-fit DML with kernel-local Welsch loss optimized via Graduated Non-Convexity and a MAD-scaled defensive OLS refit to achieve robust average dose-response estimation under localized heavy-tailed contamination while recovering outlier masks.