Inexact Nonconvex Newton-Type Methods
INFORMS Journal on Optimization · 2021-01 · 3 (2) · 154-182
DOI: 10.1287/ijoo.2019.0043 ↗Kurz gefasst
Fachartikel in der Zeitschrift INFORMS Journal on Optimization, 2021-01. Verfasst von Zhewei Yao und 3 weiteren Autor:innen. Die Arbeit wurde bisher 14-mal zitiert (Crossref) und nennt 43 Literaturangaben.
Abstract (Originalsprache)
The paper aims to extend the theory and application of nonconvex Newton-type methods, namely trust region and cubic regularization, to the settings in which, in addition to the solution of subproblems, the gradient and the Hessian of the objective function are approximated. Using certain conditions on such approximations, the paper establishes optimal worst-case iteration complexities as the exact counterparts. This paper is part of a broader research program on designing, analyzing, and implementing efficient second-order optimization methods for large-scale machine learning applications. The authors were based at UC Berkeley when the idea of the project was conceived. The first two authors were PhD students, the third author was a postdoc, all supervised by the fourth author.
Zitieren
APA
Yao, Z., Xu, P., Roosta, F., & Mahoney, M. W. (2021). Inexact Nonconvex Newton-Type Methods. INFORMS Journal on Optimization, 3(2), 154-182. https://doi.org/10.1287/ijoo.2019.0043
Vancouver
Yao Z, Xu P, Roosta F, Mahoney MW. Inexact Nonconvex Newton-Type Methods. INFORMS Journal on Optimization. 2021;3(2):154-182. doi:10.1287/ijoo.2019.0043
BibTeX
@article{Yao2021,
author = {Yao, Zhewei and Xu, Peng and Roosta, Fred and Mahoney, Michael W.},
title = {Inexact Nonconvex Newton-Type Methods},
journal = {INFORMS Journal on Optimization},
year = {2021},
volume = {3},
number = {2},
pages = {154--182},
doi = {10.1287/ijoo.2019.0043}
}
EUVIMED