A kernel-smoothed decorrelated score with cross-fitting enables valid inference for coefficients in high-dimensional classification using piecewise linear surrogate losses.
Title resolution pending
2 Pith papers cite this work. Polarity classification is still indexing.
2
Pith papers citing it
fields
stat.ME 2years
2024 2verdicts
UNVERDICTED 2representative citing papers
Reformulates ICP as multiple testing to enable FDR control with e-Closure and simultaneous true discovery bounds with closed testing, shown via simulations and US education data.
citing papers explorer
-
Inference with non-differentiable surrogate loss in a general high-dimensional classification framework
A kernel-smoothed decorrelated score with cross-fitting enables valid inference for coefficients in high-dimensional classification using piecewise linear surrogate losses.
-
On the error control of invariant causal prediction
Reformulates ICP as multiple testing to enable FDR control with e-Closure and simultaneous true discovery bounds with closed testing, shown via simulations and US education data.