2024 • Conference Paper
Convergence analysis of an inexact gradient method on smooth convex functions
Authors:
Vernimmen, Pierre ,
Glineur, François
Published in:
ESANN 2024, European Symposium on Artificial Neural Networks, Computational Intellignece and Machine Learning
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.