English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Tools annual

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Tools monthly

Global: Zendy Plus monthly

Presents the access of premium content as premium feature

Premium Content

Global: Zendy Plus annual

Global: Zendy Free Plan

This paper considers convex optimization problems where nodes of a networkhave access to summands of a global objective. Each of these local objectivesis further assumed to be an average of a finite set of functions. Themotivation for this setup is to solve large scale machine learning problemswhere elements of the training set are distributed to multiple computationalelements. The decentralized double stochastic averaging gradient (DSA)algorithm is proposed as a solution alternative that relies on: (i) The use oflocal stochastic averaging gradients. (ii) Determination of descent steps asdifferences of consecutive stochastic averaging gradients. Strong convexity oflocal functions and Lipschitz continuity of local gradients is shown toguarantee linear convergence of the sequence generated by DSA in expectation.Local iterates are further shown to approach the optimal argument for almostall realizations. The expected linear convergence of DSA is in contrast to thesublinear rate characteristic of existing methods for decentralized stochasticoptimization. Numerical experiments on a logistic regression problem illustratereductions in convergence time and number of feature vectors processed untilconvergence relative to these other alternatives.

DSA: Decentralized Double Stochastic Averaging Gradient Algorithm