Enregistré dans:
Détails bibliographiques
Auteurs principaux: Furuhashi, Takanobu, Kuroda, Hiroki, Yukawa, Masahiro, Zhao, Qibin, Hontani, Hidekata, Yokota, Tatsuya
Format: Preprint
Publié: 2026
Sujets:
Accès en ligne:https://arxiv.org/abs/2603.01304
Tags: Ajouter un tag
Pas de tags, Soyez le premier à ajouter un tag!
_version_ 1866915825954324480
author Furuhashi, Takanobu
Kuroda, Hiroki
Yukawa, Masahiro
Zhao, Qibin
Hontani, Hidekata
Yokota, Tatsuya
author_facet Furuhashi, Takanobu
Kuroda, Hiroki
Yukawa, Masahiro
Zhao, Qibin
Hontani, Hidekata
Yokota, Tatsuya
contents We propose two nonconvex regularization methods, LogLOP-l2/l1 and AdaLOP-l2/l1, for recovering block-sparse signals with unknown block partitions. These methods address the underestimation bias of existing convex approaches by extending log-sum penalty and the Minimax Concave Penalty (MCP) to the block-sparse domain via novel variational formulations. Unlike Generalized Moreau Enhancement (GME) and Bayesian methods dependent on the squared-error data fidelity term, our proposed methods are compatible with a broad range of data fidelity terms. We develop efficient Alternating Direction Method of Multipliers (ADMM)-based algorithms for these formulations that exhibit stable empirical convergence. Numerical experiments on synthetic data, angular power spectrum estimation, and denoising of nanopore currents demonstrate that our methods outperform state-of-the-art baselines in estimation accuracy.
format Preprint
id arxiv_https___arxiv_org_abs_2603_01304
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle Nonconvex Latent Optimally Partitioned Block-Sparse Recovery via Log-Sum and Minimax Concave Penalties
Furuhashi, Takanobu
Kuroda, Hiroki
Yukawa, Masahiro
Zhao, Qibin
Hontani, Hidekata
Yokota, Tatsuya
Machine Learning
We propose two nonconvex regularization methods, LogLOP-l2/l1 and AdaLOP-l2/l1, for recovering block-sparse signals with unknown block partitions. These methods address the underestimation bias of existing convex approaches by extending log-sum penalty and the Minimax Concave Penalty (MCP) to the block-sparse domain via novel variational formulations. Unlike Generalized Moreau Enhancement (GME) and Bayesian methods dependent on the squared-error data fidelity term, our proposed methods are compatible with a broad range of data fidelity terms. We develop efficient Alternating Direction Method of Multipliers (ADMM)-based algorithms for these formulations that exhibit stable empirical convergence. Numerical experiments on synthetic data, angular power spectrum estimation, and denoising of nanopore currents demonstrate that our methods outperform state-of-the-art baselines in estimation accuracy.
title Nonconvex Latent Optimally Partitioned Block-Sparse Recovery via Log-Sum and Minimax Concave Penalties
topic Machine Learning
url https://arxiv.org/abs/2603.01304