A fast non-monotone line search for stochastic gradient descent: Fid-Bau Portal

Fachinformationsdienst BAUdigital

Eine Plattform für die Wissenschaft: Bauingenieurwesen, Architektur und Urbanistik

A fast non-monotone line search for stochastic gradient descent

Fathi Hafshejani, Sajad / Gaur, Daya / Hossain, Shahadat / Benkoczi, Robert

We give an improved non-monotone line search algorithm for stochastic gradient descent (SGD) for functions that satisfy interpolation conditions. We establish theoretical convergence guarantees for the algorithm for non-convex functions. We conduct a detailed empirical evaluation to validate the theoretical results.

Zugriff

Zugriff prüfen

Verfügbarkeit in meiner Bibliothek prüfen

Bestellung bei Subito €

Seitennavigation

Dokumentinformationen

Ähnliche Titel

Exportieren, teilen und zitieren

A fast non-monotone line search for stochastic gradient descent

Fathi Hafshejani, Sajad / Gaur, Daya / Hossain, Shahadat / Benkoczi, Robert

We give an improved non-monotone line search algorithm for stochastic gradient descent (SGD) for functions that satisfy interpolation conditions. We establish theoretical convergence guarantees for the algorithm for non-convex functions. We conduct a detailed empirical evaluation to validate the theoretical results.

A fast non-monotone line search for stochastic gradient descent

Fathi Hafshejani, Sajad / Gaur, Daya / Hossain, Shahadat / Benkoczi, Robert

We give an improved non-monotone line search algorithm for stochastic gradient descent (SGD) for functions that satisfy interpolation conditions. We establish theoretical convergence guarantees for the algorithm for non-convex functions. We conduct a detailed empirical evaluation to validate the theoretical results.

Zugriff

Zugriff prüfen

Verfügbarkeit in meiner Bibliothek prüfen

Bestellung bei Subito €

Seitennavigation

Dokumentinformationen

Ähnliche Titel

Exportieren, teilen und zitieren

Dokumentinformationen

Titel:

A fast non-monotone line search for stochastic gradient descent

Weitere Titelangaben:

Optim Eng

Beteiligte:

Fathi Hafshejani, Sajad (Autor:in) / Gaur, Daya (Autor:in) / Hossain, Shahadat (Autor:in) / Benkoczi, Robert (Autor:in)

Erschienen in:

Optimization and Engineering ; 25 ; 1105-1124

Erscheinungsdatum:

01.06.2024

Format / Umfang:

20 pages

ISSN:

1573-2924 , 1389-4420

DOI:

https://doi.org/10.1007/s11081-023-09836-6

Medientyp:

Aufsatz (Zeitschrift)

Format:

Elektronische Ressource

Sprache:

Englisch

Schlagwörter:

Non-monotone line search method , Stochastic gradient descent , Non-convex problem Mathematics , Optimization , Engineering, general , Systems Theory, Control , Environmental Management , Operations Research/Decision Theory , Financial Engineering , Mathematics and Statistics

Ähnliche Titel

A fast non-monotone line search for stochastic gradient descent

Fathi Hafshejani, Sajad / Gaur, Daya / Hossain, Shahadat et al. | Springer Verlag | 2024

Digital Image Correlation Using Stochastic Parallel-Gradient-Descent Algorithm

Long, X. / Fu, S. / Qi, Z. et al. | British Library Online Contents | 2013

A linearly convergent descent method for strongly monotone variational inequality problems

Bipasha, R. S. / Yamada, K. / Yamashita, N. et al. | British Library Online Contents | 2003

A Stepwise Analytical Projected Gradient Descent Search for Hyperspectral Unmixing and Its Code Vectorization

Kizel, Fadi | Online Contents | 2017

On Early Stopping in Gradient Descent Learning

Yao, Y. / Rosasco, L. / Caponnetto, A. | British Library Online Contents | 2007