A Convergence Analysis of Approximate Message Passing with Non-Separable Functions and Applications to Multi-Class Classification

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Çakmak, Burak, Lu, Yue M., Opper, Manfred
Format: Preprint
Published: 2024
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866910329185763328
author Çakmak, Burak
Lu, Yue M.
Opper, Manfred
author_facet Çakmak, Burak
Lu, Yue M.
Opper, Manfred
contents Motivated by the recent application of approximate message passing (AMP) to the analysis of convex optimizations in multi-class classifications [Loureiro, et. al., 2021], we present a convergence analysis of AMP dynamics with non-separable multivariate nonlinearities. As an application, we present a complete (and independent) analysis of the motivated convex optimization problem.
format Preprint
id arxiv_https___arxiv_org_abs_2402_08676
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle A Convergence Analysis of Approximate Message Passing with Non-Separable Functions and Applications to Multi-Class Classification
Çakmak, Burak
Lu, Yue M.
Opper, Manfred
Machine Learning
Information Theory
Motivated by the recent application of approximate message passing (AMP) to the analysis of convex optimizations in multi-class classifications [Loureiro, et. al., 2021], we present a convergence analysis of AMP dynamics with non-separable multivariate nonlinearities. As an application, we present a complete (and independent) analysis of the motivated convex optimization problem.
title A Convergence Analysis of Approximate Message Passing with Non-Separable Functions and Applications to Multi-Class Classification
topic Machine Learning
Information Theory
url https://arxiv.org/abs/2402.08676