A Convergence Analysis of Approximate Message Passing with Non-Separable Functions and Applications to Multi-Class Classification
Machine Learning
2024-02-14 v1 Information Theory
math.IT
Abstract
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.
Cite
@article{arxiv.2402.08676,
title = {A Convergence Analysis of Approximate Message Passing with Non-Separable Functions and Applications to Multi-Class Classification},
author = {Burak Çakmak and Yue M. Lu and Manfred Opper},
journal= {arXiv preprint arXiv:2402.08676},
year = {2024}
}