Salvato in:
Dettagli Bibliografici
Autori principali: Chen, Hengchao, Chen, Xin, Elmasri, Mohamad, Sun, Qiang
Natura: Preprint
Pubblicazione: 2023
Soggetti:
Accesso online:https://arxiv.org/abs/2305.19206
Tags: Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
_version_ 1866910466271346688
author Chen, Hengchao
Chen, Xin
Elmasri, Mohamad
Sun, Qiang
author_facet Chen, Hengchao
Chen, Xin
Elmasri, Mohamad
Sun, Qiang
contents Gradient Descent (GD) has been proven effective in solving various matrix factorization problems. However, its optimization behavior with large initial values remains less understood. To address this gap, this paper presents a novel theoretical framework for examining the convergence trajectory of GD with a large initialization. The framework is grounded in signal-to-noise ratio concepts and inductive arguments. The results uncover an implicit incremental learning phenomenon in GD and offer a deeper understanding of its performance in large initialization scenarios.
format Preprint
id arxiv_https___arxiv_org_abs_2305_19206
institution arXiv
publishDate 2023
record_format arxiv
spellingShingle Gradient descent in matrix factorization: Understanding large initialization
Chen, Hengchao
Chen, Xin
Elmasri, Mohamad
Sun, Qiang
Optimization and Control
Machine Learning
Gradient Descent (GD) has been proven effective in solving various matrix factorization problems. However, its optimization behavior with large initial values remains less understood. To address this gap, this paper presents a novel theoretical framework for examining the convergence trajectory of GD with a large initialization. The framework is grounded in signal-to-noise ratio concepts and inductive arguments. The results uncover an implicit incremental learning phenomenon in GD and offer a deeper understanding of its performance in large initialization scenarios.
title Gradient descent in matrix factorization: Understanding large initialization
topic Optimization and Control
Machine Learning
url https://arxiv.org/abs/2305.19206