Nonparametric linear feature learning in regression through regularisation

• Published in: Electronic journal of statistics Vol. 18; no. 2
• Main Authors: Follain, Bertille, Bach, Francis
• Format: Journal Article
• Language: English
• Published: Shaker Heights, OH : Institute of Mathematical Statistics 01.01.2024
• Subjects: Artificial Intelligence; Computer Science; Machine Learning; Methodology; Statistics; Statistics Theory; non-parametric regression; secondary 65K10 multi-index model sparsity non-parametric regression regularised empirical risk minimisation alternating minimisation Hermite polynomials; secondary 65K10 multi-index model; Regularised empirical risk minimisation; Nonparametric regression; MSC2020 subject classifications: Primary 62G08; MSC2020 subject classifications: Primary 62G08 62F10; Alternating minimisation; Sparsity; Multi-index model; 62F10; Hermite orthogonal polynomials; Hermite polynomials

Representation learning plays a crucial role in automated feature selection, particularly in the context of high-dimensional data, where non-parametric methods often struggle. In this study, we focus on supervised learning scenarios where the pertinent information resides within a lower-dimensional...