Inference with non-differentiable surrogate loss in a general high-dimensional classification framework

Дата публикации: 17-08-2026 20:26:00

Penalized empirical risk minimization with a surrogate loss function is often used to learn a high-dimensional linear decision rule in classification problems. Although much of the literature focus on the generalization error, there is a lack of inference procedures for identifying the driving factors of the estimated decision rule, especially when the surrogate loss is non-differentiable. We propose a kernel-smoothed decorrelated score to construct hypothesis tests and interval estimators for a linear decision rule estimated using a piece-wise linear surrogate loss, which has a discontinuous gradient and non-regular Hessian. Specifically, we adopt kernel approximations to smooth the discontinuous gradient near discontinuity points and approximate the non-regular Hessian of the surrogate loss. In applications where additional nuisance parameters are involved, we propose a novel cross-fitted version to accommodate flexible nuisance estimates and kernel approximations. We establish the limiting distribution of the kernel-smoothed decorrelated score and its cross-fitted version in a high-dimensional setup. Simulation and real data analysis are conducted to demonstrate the validity and the superiority of the proposed method.

Схожие новости

#	Наименование новости	Тональность	Информативность	Дата публикации
1	Refined Risk Bounds for Unbounded Losses via Transductive Priors	0	5.33	17-08-2026
2	The Distribution of Ridgeless Least Squares Interpolators	0	5.78	17-08-2026
3	Near-optimal Delta-convex Estimation of Lipschitz Functions	0	9.71	17-08-2026
4	Decorrelated Local Linear Estimator: Inference for Non-linear Effects in High-dimensional Additive Models	0	7.75	17-08-2026
5	Transfer Learning via Regularized Random-effects Linear Discriminant Analysis	0	5.98	17-08-2026
6	Knowledge Cascade: Reverse Knowledge Distillation on Nonparametric Multivariate Functional Estimation	0	5.75	17-08-2026
7	Semi-supervised learning for linear extremile regression	0	6.83	17-08-2026
8	Deconvolution in unlinked linear models	0	4.62	17-08-2026
9	High-dimensional Parameter Transfer With Fused-Regularizer	0	9.38	17-08-2026
10	Differentially Private Estimation and Inference in High-Dimensional Regression with FDR Control	0	7.94	17-08-2026

Классификация: . Схожих патентов: 0. Схожих новостей: 10. Тональность: 0. Информативность: 8.38. Источник: jmlr.org.