Convergence analysis of an inexact gradient method on smooth convex functions

Vernimmen, Pierre;Glineur, François
(2024) ESANN 2024, European Symposium on Artificial Neural Networks, Computational Intellignece and Machine Learning — Location: Bruges (Belgium) and online (9.October.2024)

Files

ES2024-171.pdf
  • Open Access
  • Adobe PDF
  • 1.56 MB
Download

Details

Authors
Abstract
We consider the classical gradient method with constant stepsizes where some error is introduced in the computation of each gradient. More specifically, we assume relative inexactness, in the sense that the norm of the difference between the true gradient and its approximate value is bounded by a certain fraction of the gradient norm. We establish a sublinear convergence rate for this inexact method when applied to smooth convex functions, and illustrate on a logistic regression example.
Affiliations

Citations

  • APA
  • Chicago
  • FWB

Vernimmen, P., & Glineur, F. (2024). Convergence analysis of an inexact gradient method on smooth convex functions. ESANN 2024 proceedings. Published. ESANN 2024, European Symposium on Artificial Neural Networks, Computational Intellignece and Machine Learning, Bruges (Belgium) and online. https://doi.org/10.14428/esann/2024.es2024-171