AI Paper Atlas

Research map / Optimization / Distributed and compressed

DSA: Decentralized Double Stochastic Averaging Gradient Algorithm

Aryan Mokhtari, Alejandro Ribeiro

JMLR 2016 · Optimization › Distributed and compressed · approach: Decentralized optimization algorithms · 141 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (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
use DSA instead of EXTRA (fewer gradients per iteration)EXTRA · logistic regression Q=500, N=20gradient evaluations to error 1e-8: 1725 -> 380Improved
use DSA instead of EXTRAEXTRA · logistic regression Q=500, N=20iterations to error 1e-8: 69 -> 380Hurt
add double descent (EXTRA mixing) to stochastic averaging gradient (ablation)Decentralized SAGA · logistic regression Q=500, N=20convergenceImproved
add stochastic averaging to EXTRA instead of plain stochastic gradient (ablation)Stochastic EXTRA · logistic regression Q=500, N=20convergenceImproved
reduce step size from 1e-2 to 1e-3 (ablation)DGD, sto-EXTRA, D-SAGA · logistic regression Q=500, N=20accuracy vs speed: alpha = 10^-2 -> 10^-3Mixed
use complete graph instead of line graph (ablation)DSA · N=50, Q=500iterations to error 1e-8: 1819 -> 247Improved
use cycle graph instead of line graph (ablation)DSA · N=50, Q=500iterations to error 1e-8: 1819 -> 1133Improved
use random graph pc=0.35 instead of pc=0.25 (ablation)DSA · N=50, Q=500iterations to error 1e-8: 504 -> 310Improved
increase total samples Q from 100 to 5000 (q from 5 to 250) (ablation)DSA · N=20 random graphconvergence timeHurt
larger graph condition number (ablation)DSA vs EXTRA · N=50 topologiesDSA advantage over EXTRAImproved
choose N=125 (q=4) nodes instead of very small or large N (ablation)DSA · Q=500convergenceImproved
use DSA instead of EXTRA on large problemEXTRA · protein homology, Q=1.45x10^5, N=200gradient evaluations to target accuracyImproved

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