Simultaneous perturbation stochastic approximation: towards one-measurement per iteration
Numerical Algorithms · 2023-11 · 94 (3) · 1085-1101
DOI: 10.1007/s11075-023-01528-7 ↗Kurz gefasst
Fachartikel in der Zeitschrift Numerical Algorithms, online erschienen am 15.05.2023. Verfasst von Shiru Li und 2 weiteren Autor:innen. Die Arbeit wurde bisher 5-mal zitiert (Crossref) und nennt 18 Literaturangaben. Gefördert von: National Natural Science Foundation of China.
Abstract (Originalsprache)
When measuring the value of a function to be minimized is not only expensive but also with noise, the popular simultaneous perturbation stochastic approximation (SPSA) algorithm requires only two function values in each iteration. In this paper, we present a method requiring only one function measurement value per iteration in the average sense. We prove the strong convergence and asymptotic normality of the new algorithm. Limited experimental results demonstrate the effectiveness and potential of our algorithm for solving low-dimensional problems.
Förderung
- National Natural Science Foundation of China
Zitieren
APA
Li, S., Xia, Y., & Xu, Z. (2023). Simultaneous perturbation stochastic approximation: towards one-measurement per iteration. Numerical Algorithms, 94(3), 1085-1101. https://doi.org/10.1007/s11075-023-01528-7
Vancouver
Li S, Xia Y, Xu Z. Simultaneous perturbation stochastic approximation: towards one-measurement per iteration. Numerical Algorithms. 2023;94(3):1085-1101. doi:10.1007/s11075-023-01528-7
BibTeX
@article{Li2023,
author = {Li, Shiru and Xia, Yong and Xu, Zi},
title = {Simultaneous perturbation stochastic approximation: towards one-measurement per iteration},
journal = {Numerical Algorithms},
year = {2023},
volume = {94},
number = {3},
pages = {1085--1101},
doi = {10.1007/s11075-023-01528-7}
}
EUVIMED