English

Statistical Inference in Classification of High-dimensional Gaussian Mixture

Machine Learning 2024-10-29 v1 Machine Learning

Abstract

We consider the classification problem of a high-dimensional mixture of two Gaussians with general covariance matrices. Using the replica method from statistical physics, we investigate the asymptotic behavior of a general class of regularized convex classifiers in the high-dimensional limit, where both the sample size nn and the dimension pp approach infinity while their ratio α=n/p\alpha=n/p remains fixed. Our focus is on the generalization error and variable selection properties of the estimators. Specifically, based on the distributional limit of the classifier, we construct a de-biased estimator to perform variable selection through an appropriate hypothesis testing procedure. Using L1L_1-regularized logistic regression as an example, we conducted extensive computational experiments to confirm that our analytical findings are consistent with numerical simulations in finite-sized systems. We also explore the influence of the covariance structure on the performance of the de-biased estimator.

Keywords

Cite

@article{arxiv.2410.19950,
  title  = {Statistical Inference in Classification of High-dimensional Gaussian Mixture},
  author = {Hanwen Huang and Peng Zeng},
  journal= {arXiv preprint arXiv:2410.19950},
  year   = {2024}
}

Comments

22 pages, 4 figures

R2 v1 2026-06-28T19:36:11.403Z