We gratefully acknowledge support from
the Simons Foundation and member institutions.
Full-text links:

Download:

Current browse context:

stat.ML

Change to browse by:

References & Citations

Bookmark

(what is this?)
CiteULike logo BibSonomy logo Mendeley logo del.icio.us logo Digg logo Reddit logo ScienceWISE logo

Statistics > Machine Learning

Title: Stochastic gradient descent with noise of machine learning type. Part I: Discrete time analysis

Abstract: Stochastic gradient descent (SGD) is one of the most popular algorithms in modern machine learning. The noise encountered in these applications is different from that in many theoretical analyses of stochastic gradient algorithms. In this article, we discuss some of the common properties of energy landscapes and stochastic noise encountered in machine learning problems, and how they affect SGD-based optimization.
In particular, we show that the learning rate in SGD with machine learning noise can be chosen to be small, but uniformly positive for all times if the energy landscape resembles that of overparametrized deep learning problems. If the objective function satisfies a Lojasiewicz inequality, SGD converges to the global minimum exponentially fast, and even for functions which may have local minima, we establish almost sure convergence to the global minimum at an exponential rate from any finite energy initialization. The assumptions that we make in this result concern the behavior where the objective function is either small or large and the nature of the gradient noise, but the energy landscape is fairly unconstrained on the domain where the objective function takes values in an intermediate regime.
Subjects: Machine Learning (stat.ML); Machine Learning (cs.LG); Optimization and Control (math.OC)
MSC classes: Primary: 90C26, 90C15. Secondary: 68T07, 90C30
Cite as: arXiv:2105.01650 [stat.ML]
  (or arXiv:2105.01650v1 [stat.ML] for this version)

Submission history

From: Stephan Wojtowytsch [view email]
[v1] Tue, 4 May 2021 17:52:20 GMT (1147kb,D)

Link back to: arXiv, form interface, contact.