Exploiting Negative Curvature in Conjunction with Adaptive Sampling: Theoretical Results and a Practical Algorithm

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Berahas, Albert S., Bollapragada, Raghu, Dong, Wanping
Format: Preprint
Published: 2024
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866910699890933760
author Berahas, Albert S.
Bollapragada, Raghu
Dong, Wanping
author_facet Berahas, Albert S.
Bollapragada, Raghu
Dong, Wanping
contents In this paper, we propose algorithms that exploit negative curvature for solving noisy nonlinear nonconvex unconstrained optimization problems. We consider both deterministic and stochastic inexact settings, and develop two-step algorithms that combine directions of negative curvature and descent directions to update the iterates. Under reasonable assumptions, we prove second-order convergence results and derive complexity guarantees for both settings. To tackle large-scale problems, we develop a practical variant that utilizes the conjugate gradient method with negative curvature detection and early stopping to compute a step, a simple adaptive step size scheme, and a strategy for selecting the sample sizes of the gradient and Hessian approximations as the optimization progresses. Numerical results on two machine learning problems showcase the efficacy and efficiency of the practical method.
format Preprint
id arxiv_https___arxiv_org_abs_2411_10378
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Exploiting Negative Curvature in Conjunction with Adaptive Sampling: Theoretical Results and a Practical Algorithm
Berahas, Albert S.
Bollapragada, Raghu
Dong, Wanping
Optimization and Control
In this paper, we propose algorithms that exploit negative curvature for solving noisy nonlinear nonconvex unconstrained optimization problems. We consider both deterministic and stochastic inexact settings, and develop two-step algorithms that combine directions of negative curvature and descent directions to update the iterates. Under reasonable assumptions, we prove second-order convergence results and derive complexity guarantees for both settings. To tackle large-scale problems, we develop a practical variant that utilizes the conjugate gradient method with negative curvature detection and early stopping to compute a step, a simple adaptive step size scheme, and a strategy for selecting the sample sizes of the gradient and Hessian approximations as the optimization progresses. Numerical results on two machine learning problems showcase the efficacy and efficiency of the practical method.
title Exploiting Negative Curvature in Conjunction with Adaptive Sampling: Theoretical Results and a Practical Algorithm
topic Optimization and Control
url https://arxiv.org/abs/2411.10378