AI-SARAH: Adaptive and Implicit Stochastic Recursive Gradient Methods
Machine Learning
2022-02-02 v3 Optimization and Control
Abstract
We present AI-SARAH, a practical variant of SARAH. As a variant of SARAH, this algorithm employs the stochastic recursive gradient yet adjusts step-size based on local geometry. AI-SARAH implicitly computes step-size and efficiently estimates local Lipschitz smoothness of stochastic functions. It is fully adaptive, tune-free, straightforward to implement, and computationally efficient. We provide technical insight and intuitive illustrations on its design and convergence. We conduct extensive empirical analysis and demonstrate its strong performance compared with its classical counterparts and other state-of-the-art first-order methods in solving convex machine learning problems.
Cite
@article{arxiv.2102.09700,
title = {AI-SARAH: Adaptive and Implicit Stochastic Recursive Gradient Methods},
author = {Zheng Shi and Abdurakhmon Sadiev and Nicolas Loizou and Peter Richtárik and Martin Takáč},
journal= {arXiv preprint arXiv:2102.09700},
year = {2022}
}