DML estimators for the quadratic functional and quadratic density integral are asymptotically inadmissible under SA models and dominated by empirical HOIF estimators, while DML remains minimax for expected conditional covariance.
Sharp structure-agnostic lower bounds for general functional estimation
3 Pith papers cite this work. Polarity classification is still indexing.
3
Pith papers citing it
years
2026 3verdicts
UNVERDICTED 3representative citing papers
Proposes a new estimator for β0 in the partial linear model that attains rate n^{-1/2} + δ^a_μ + (δ^s_μ)^2 with matching lower bound, eliminating first-order stochastic nuisance error.
Develops higher-order influence function estimators for implicitly defined parameters in non-separable structural models using U-processes theory.
citing papers explorer
-
On the Asymptotic Inadmissibility of Double Machine Learning Estimators Under Structure-Agnostic Models
DML estimators for the quadratic functional and quadratic density integral are asymptotically inadmissible under SA models and dominated by empirical HOIF estimators, while DML remains minimax for expected conditional covariance.
-
Optimally taming biases in black-box models for efficient semiparametric estimation
Proposes a new estimator for β0 in the partial linear model that attains rate n^{-1/2} + δ^a_μ + (δ^s_μ)^2 with matching lower bound, eliminating first-order stochastic nuisance error.