pith. sign in

arxiv: 1902.05679 · v2 · pith:7POWAYZYnew · submitted 2019-02-15 · 🧮 math.OC · cs.LG· stat.ML

ProxSARAH: An Efficient Algorithmic Framework for Stochastic Composite Nonconvex Optimization

classification 🧮 math.OC cs.LGstat.ML
keywords algorithmsnonconvexboundscomplexitycompositeconstantexistingproblems
0
0 comments X
read the original abstract

We propose a new stochastic first-order algorithmic framework to solve stochastic composite nonconvex optimization problems that covers both finite-sum and expectation settings. Our algorithms rely on the SARAH estimator introduced in (Nguyen et al, 2017) and consist of two steps: a proximal gradient and an averaging step making them different from existing nonconvex proximal-type algorithms. The algorithms only require an average smoothness assumption of the nonconvex objective term and additional bounded variance assumption if applied to expectation problems. They work with both constant and adaptive step-sizes, while allowing single sample and mini-batches. In all these cases, we prove that our algorithms can achieve the best-known complexity bounds. One key step of our methods is new constant and adaptive step-sizes that help to achieve desired complexity bounds while improving practical performance. Our constant step-size is much larger than existing methods including proximal SVRG schemes in the single sample case. We also specify the algorithm to the non-composite case that covers existing state-of-the-arts in terms of complexity bounds. Our update also allows one to trade-off between step-sizes and mini-batch sizes to improve performance. We test the proposed algorithms on two composite nonconvex problems and neural networks using several well-known datasets.

This paper has not been read by Pith yet.

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Forward citations

Cited by 2 Pith papers

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score.

  1. A Stochastic Composite Gradient Method with Incremental Variance Reduction

    math.OC 2019-06 unverdicted novelty 6.0

    Proposes an incremental variance-reduced stochastic gradient method for minimizing smooth nonconvex composite functions that achieves optimal first-order complexity rates.

  2. Accelerating Mini-batch SARAH by Step Size Rules

    cs.LG 2019-06 unverdicted novelty 4.0

    MB-SARAH-RBB uses a random Barzilai-Borwein step size to accelerate mini-batch SARAH, with a linear convergence proof and improved complexity for strongly convex objectives.