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 growing amount of high dimensional data in different machine learningapplications requires more efficient and scalable optimization algorithms. Inthis work, we consider combining two techniques, parallelism and Nesterov'sacceleration, to design faster algorithms for L1-regularized loss. We firstsimplify BOOM, a variant of gradient descent, and study it in a unifiedframework, which allows us to not only propose a refined measurement ofsparsity to improve BOOM, but also show that BOOM is provably slower thanFISTA. Moving on to parallel coordinate descent methods, we then propose anefficient accelerated version of Shotgun, improving the convergence rate from$O(1/t)$ to $O(1/t^2)$. Our algorithm enjoys a concise form and analysiscompared to previous work, and also allows one to study several connected workin a unified way.

Accelerated Parallel Optimization Methods for Large Scale Machine Learning