atlas

Research map / Optimization / Distributed and compressed

A Delay-tolerant Proximal-Gradient Algorithm for Distributed Learning

ICML 2018 · Optimization › Distributed and compressed · approach: Decentralized optimization algorithms · 20 citations (OpenAlex)

Read the paperCheck your idea against 108,967 papers

AI-extracted summary. Written by AI from the paper's title only (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
Asynchronous distributed averaging with repeated local proximal-gradient steps (DAve-RPG) versus synchronous proximal gradient and PIAGdistributed regularized logistic regression · RCV1 and URL (wall clock)suboptimality vs wall clock timeHelped
Increasing local inner proximal-gradient iterations p from 1 to 7 (ablation)DAve-RPG · URL, 100 workerstime to reach minimumHelped
Increasing local inner iterations to p=10 (ablation)DAve-RPG · URL, 100 workersconvergence speedHurt
Adding large simulated random machine delays and crashes (delays up to about 300 iterations) (ablation)DAve-RPG · RCV1, 10 machinessuboptimalityNo clear effect
Scaling number of workers (speedup measured as inverse time to suboptimality 1e-2) (ablation)DAve-RPG with p=50 · RCV1, up to 100 machinesspeedupMixed

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy