Saved in:
Bibliographic Details
Main Authors: Chakrabarti, Kushal, Baranwal, Mayank
Format: Preprint
Published: 2024
Subjects:
Online Access:https://arxiv.org/abs/2407.12629
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866929424703684608
author Chakrabarti, Kushal
Baranwal, Mayank
author_facet Chakrabarti, Kushal
Baranwal, Mayank
contents Adaptive gradient-descent optimizers are the standard choice for training neural network models. Despite their faster convergence than gradient-descent and remarkable performance in practice, the adaptive optimizers are not as well understood as vanilla gradient-descent. A reason is that the dynamic update of the learning rate that helps in faster convergence of these methods also makes their analysis intricate. Particularly, the simple gradient-descent method converges at a linear rate for a class of optimization problems, whereas the practically faster adaptive gradient methods lack such a theoretical guarantee. The Polyak-Łojasiewicz (PL) inequality is the weakest known class, for which linear convergence of gradient-descent and its momentum variants has been proved. Therefore, in this paper, we prove that AdaGrad and Adam, two well-known adaptive gradient methods, converge linearly when the cost function is smooth and satisfies the PL inequality. Our theoretical framework follows a simple and unified approach, applicable to both batch and stochastic gradients, which can potentially be utilized in analyzing linear convergence of other variants of Adam.
format Preprint
id arxiv_https___arxiv_org_abs_2407_12629
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle A Methodology Establishing Linear Convergence of Adaptive Gradient Methods under PL Inequality
Chakrabarti, Kushal
Baranwal, Mayank
Machine Learning
Artificial Intelligence
Optimization and Control
Adaptive gradient-descent optimizers are the standard choice for training neural network models. Despite their faster convergence than gradient-descent and remarkable performance in practice, the adaptive optimizers are not as well understood as vanilla gradient-descent. A reason is that the dynamic update of the learning rate that helps in faster convergence of these methods also makes their analysis intricate. Particularly, the simple gradient-descent method converges at a linear rate for a class of optimization problems, whereas the practically faster adaptive gradient methods lack such a theoretical guarantee. The Polyak-Łojasiewicz (PL) inequality is the weakest known class, for which linear convergence of gradient-descent and its momentum variants has been proved. Therefore, in this paper, we prove that AdaGrad and Adam, two well-known adaptive gradient methods, converge linearly when the cost function is smooth and satisfies the PL inequality. Our theoretical framework follows a simple and unified approach, applicable to both batch and stochastic gradients, which can potentially be utilized in analyzing linear convergence of other variants of Adam.
title A Methodology Establishing Linear Convergence of Adaptive Gradient Methods under PL Inequality
topic Machine Learning
Artificial Intelligence
Optimization and Control
url https://arxiv.org/abs/2407.12629