Current browse context:
math.ST
Change to browse by:
References & Citations
Mathematics > Statistics Theory
Title: Post-selection inference with HSIC-Lasso
(Submitted on 29 Oct 2020 (v1), last revised 17 Jun 2021 (this version, v2))
Abstract: Detecting influential features in non-linear and/or high-dimensional data is a challenging and increasingly important task in machine learning. Variable selection methods have thus been gaining much attention as well as post-selection inference. Indeed, the selected features can be significantly flawed when the selection procedure is not accounted for. We propose a selective inference procedure using the so-called model-free "HSIC-Lasso" based on the framework of truncated Gaussians combined with the polyhedral lemma. We then develop an algorithm, which allows for low computational costs and provides a selection of the regularisation parameter. The performance of our method is illustrated by both artificial and real-world data based experiments, which emphasise a tight control of the type-I error, even for small sample sizes.
Submission history
From: Tobias Freidling [view email][v1] Thu, 29 Oct 2020 15:10:21 GMT (151kb)
[v2] Thu, 17 Jun 2021 11:19:29 GMT (804kb,D)
Link back to: arXiv, form interface, contact.