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 Plus annual

Presents the access of premium content as premium feature

Premium Content

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 Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

In this article we establish new central limit theorems for Ruppert-Polyakaveraged stochastic gradient descent schemes. Compared to previous work we donot assume that convergence occurs to an isolated attractor but instead allowconvergence to a stable manifold. On the stable manifold the target function isconstant and the oscillations in the tangential direction may be significantlylarger than the ones in the normal direction. As we show, one still recovers acentral limit theorem with the same rates as in the case of isolatedattractors. Here we consider step-sizes $\gamma_n=n^{-\gamma}$ with$\gamma\in(\frac34,1)$, typically.

Central limit theorems for stochastic gradient descent with averaging for stable manifolds