Fast Incremental Method for Nonconvex Optimization
Optimization and Control
2016-03-22 v1 Machine Learning
Machine Learning
Abstract
We analyze a fast incremental aggregated gradient method for optimizing nonconvex problems of the form . Specifically, we analyze the SAGA algorithm within an Incremental First-order Oracle framework, and show that it converges to a stationary point provably faster than both gradient descent and stochastic gradient descent. We also discuss a Polyak's special class of nonconvex problems for which SAGA converges at a linear rate to the global optimum. Finally, we analyze the practically valuable regularized and minibatch variants of SAGA. To our knowledge, this paper presents the first analysis of fast convergence for an incremental aggregated gradient method for nonconvex problems.
Cite
@article{arxiv.1603.06159,
title = {Fast Incremental Method for Nonconvex Optimization},
author = {Sashank J. Reddi and Suvrit Sra and Barnabas Poczos and Alex Smola},
journal= {arXiv preprint arXiv:1603.06159},
year = {2016}
}