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

The Frank-Wolfe (FW) method is a popular approach for solving optimizationproblems with structured constraints that arise in machine learningapplications. In recent years, stochastic versions of FW have gainedpopularity, motivated by large datasets for which the computation of the fullgradient is prohibitively expensive. In this paper, we present two new variantsof the FW algorithms for stochastic finite-sum minimization. Our algorithmshave the best convergence guarantees of existing stochastic FW approaches forboth convex and non-convex objective functions. Our methods do not have theissue of permanently collecting large batches, which is common to manystochastic projection-free approaches. Moreover, our second approach does notrequire either large batches or full deterministic gradients, which is atypical weakness of many techniques for finite-sum problems. The fastertheoretical rates of our approaches are confirmed experimentally.

Sarah Frank-Wolfe: Methods for Constrained Optimization with Best Rates and Practical Features