Salvato in:
Dettagli Bibliografici
Autori principali: Zanco, Daniel Gomes de Pinho, Szczecinski, Leszek, Benesty, Jacob, Kuhn, Eduardo Vinicius
Natura: Preprint
Pubblicazione: 2025
Soggetti:
Accesso online:https://arxiv.org/abs/2512.14932
Tags: Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
_version_ 1866908717638746112
author Zanco, Daniel Gomes de Pinho
Szczecinski, Leszek
Benesty, Jacob
Kuhn, Eduardo Vinicius
author_facet Zanco, Daniel Gomes de Pinho
Szczecinski, Leszek
Benesty, Jacob
Kuhn, Eduardo Vinicius
contents In this work, we propose a method to efficiently find the regularization parameter for low-rank MMSE filters based on a Kronecker-product representation. We show that the regularization parameter is surprisingly linked to the problem of rank selection and, thus, properly choosing it, is crucial for low-rank settings. The proposed method is validated through simulations, showing significant gains over commonly used methods.
format Preprint
id arxiv_https___arxiv_org_abs_2512_14932
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Low-rank MMSE filters, Kronecker-product representation, and regularization: a new perspective
Zanco, Daniel Gomes de Pinho
Szczecinski, Leszek
Benesty, Jacob
Kuhn, Eduardo Vinicius
Machine Learning
In this work, we propose a method to efficiently find the regularization parameter for low-rank MMSE filters based on a Kronecker-product representation. We show that the regularization parameter is surprisingly linked to the problem of rank selection and, thus, properly choosing it, is crucial for low-rank settings. The proposed method is validated through simulations, showing significant gains over commonly used methods.
title Low-rank MMSE filters, Kronecker-product representation, and regularization: a new perspective
topic Machine Learning
url https://arxiv.org/abs/2512.14932