cs.LGDec 16, 2025

Automatic regularization for Kronecker product MMSE filters

Authors: Daniel Gomes de Pinho Zanco, Leszek Szczecinski, Jacob Benesty, Eduardo Vinicius Kuhn

Organizations: INRS–Institut National de la Recherche Scientific, Montreal, QC, H5A-1K6, Canada. · Federal University of Technology - Paraná, Toledo, Paraná, 85902-490, Brazil.

Abstract

In this work, we propose a method to automatically find the regularization parameter for low-rank MMSE filters based on a Kronecker-product representation, using an approximate leave-one-out (ALO) cross-validation metric. We formulate the problem in a constrained form that covers both error minimization and interference suppression, and explore the link between regularization and rank selection. The proposed method is validated on system identification and antenna beamforming problems, showing significant gains over commonly used methods.

Figures & tables

Explore similar work

CardsList
  1. L1 Regularization Paths in Linear Models by Parametric Gaussian Message Passing

    Apr 18, 2026Yun-Peng Li, Hans-Andrea LoeligerLinear Regression

  2. Online Algorithms for Independent Low-Rank Matrix Analysis and Rank-Constrained Spatial Covariance Matrix Estimation Based on Maximum Weighted Likelihood Estimation

    Sep 18, 2026Yuto Ishikawa, Norihiro Takamune, Tomohiko Nakamura +4Audio Source SeparationCovariance Estimation

  3. Two-Layer Linear Auto-Regressive Models Estimate Latent States

    Jun 10, 2026Yahya Sattar, Sunmook Choi, Leo Maynard-Zhang +3Dynamical SystemsKalman Filtering