A note on diffusion limits for stochastic gradient descent

DOI10.1016/J.JAT.2025.106160MaRDI QIDQ6993577zbMATH OpenFDO

Authors A. Lanconelli, Christopher S. A. Lauria

Publication date 25 April 2025

Published in Journal of Approximation Theory (Search for Journal in Brave)

zbMATH Keywords

Markov processstochastic differential equationBrownian motionstochastic gradient descent

Mathematics Subject Classification ID

Nonconvex programming, global optimization (90C26) Stochastic programming (90C15) Computational methods for stochastic equations (aspects of stochastic analysis) (60H35) Derivative-free methods and methods using generalized derivatives (90C56)

Cites work

• scientific article; zbMATH DE number 5913352 (Why is no real title available?)
• scientific article; zbMATH DE number 1569102 (Why is no real title available?)
• scientific article; zbMATH DE number 7306890 (Why is no real title available?)
• A Stochastic Approximation Method
• ARCH models as diffusion approximations
• Adaption and learning in automatic systems. Translated by Z. J. Nikolic
• Adaptive subgradient methods for online learning and stochastic optimization
• Asymptotic and finite-sample properties of estimators based on stochastic gradients
• Learning representations by back-propagating errors
• Multidimensional diffusion processes.
• Optimization methods for large-scale machine learning
• Reconsidering the continuous time limit of the GARCH(1,1) process
• Robust Stochastic Approximation Approach to Stochastic Programming
• Stochastic modified equations and dynamics of stochastic gradient algorithms. I: Mathematical foundations
• The continuous-time limit of score-driven volatility models

This page was built for publication: A note on diffusion limits for stochastic gradient descent

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q6993577)