Saved in:
Bibliographic Details
Main Authors: Nagano, Yasushi, Hukushima, Koji
Format: Preprint
Published: 2023
Subjects:
Online Access:https://arxiv.org/abs/2306.02607
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866912256278659072
author Nagano, Yasushi
Hukushima, Koji
author_facet Nagano, Yasushi
Hukushima, Koji
contents In sparse signal processing, this study investigates the effect of the global shrinkage parameter $τ$ of a horseshoe prior, one of the global-local shrinkage prior, on the linear regression. Statistical mechanics methods are employed to examine the accuracy of signal estimation. A phase diagram of successful and failure of signal recovery in noise-less compressed sensing with varying $τ$ is discussed from the viewpoint of dynamic characterization of the approximate message passing as a solving algorithm and static characterization of the free-energy landscape. It is found that there exists a parameter region where the approximate message passing algorithm can hardly recover the true signal, even though the true signal is locally stable. The analysis of the free-energy landscape also provides important insight into the optimal choice of $τ$.
format Preprint
id arxiv_https___arxiv_org_abs_2306_02607
institution arXiv
publishDate 2023
record_format arxiv
spellingShingle Effect of global shrinkage parameter of horseshoe prior in compressed sensing
Nagano, Yasushi
Hukushima, Koji
Disordered Systems and Neural Networks
In sparse signal processing, this study investigates the effect of the global shrinkage parameter $τ$ of a horseshoe prior, one of the global-local shrinkage prior, on the linear regression. Statistical mechanics methods are employed to examine the accuracy of signal estimation. A phase diagram of successful and failure of signal recovery in noise-less compressed sensing with varying $τ$ is discussed from the viewpoint of dynamic characterization of the approximate message passing as a solving algorithm and static characterization of the free-energy landscape. It is found that there exists a parameter region where the approximate message passing algorithm can hardly recover the true signal, even though the true signal is locally stable. The analysis of the free-energy landscape also provides important insight into the optimal choice of $τ$.
title Effect of global shrinkage parameter of horseshoe prior in compressed sensing
topic Disordered Systems and Neural Networks
url https://arxiv.org/abs/2306.02607