Sequential convergence of AdaGrad algorithm for smooth convex optimization
From MaRDI portal
Publication:6354489
DOI10.1016/J.ORL.2021.04.011arXiv2011.12341MaRDI QIDQ6354489FDOQ6354489
Authors: Cheik Traoré, Edouard Pauwels 
Publication date: 24 November 2020
Abstract: We prove that the iterates produced by, either the scalar step size variant, or the coordinatewise variant of AdaGrad algorithm, are convergent sequences when applied to convex objective functions with Lipschitz gradient. The key insight is to remark that such AdaGrad sequences satisfy a variable metric quasi-Fej'er monotonicity property, which allows to prove convergence.
This page was built for publication: Sequential convergence of AdaGrad algorithm for smooth convex optimization
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q6354489)
