AI Paper Atlas

Research map / Optimization / Nonconvex and smooth optimization

On the Influence of Momentum Acceleration on Online Learning

Kun Yuan, Bicheng Ying, Ali H. Sayed

JMLR 2016 · Optimization › Nonconvex and smooth optimization · approach: Stochastic gradient last-iterate analysis · 31 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 momentum (beta=0.9) with the same step-size 0.003 instead of plain LMSLMS · Synthetic LMS, 10-dimMSD: mu=mu_m=0.003, beta=0.9Mixed
scale momentum step-size down to mu_m=mu(1-beta)=0.0003Momentum LMS · Synthetic LMS, 10-dimMSD curve: mu_m=0.0003 vs mu=0.003Improved
add decaying momentum parameter beta(i)Momentum LMS · Synthetic LMS, 10-dimconvergence and steady-state MSDImproved
scale momentum step-size to mu_m=mu(1-beta)=0.0005Momentum SGD · Synthetic logistic regressionMSD curve: mu_m=0.0005 vs mu=0.005Improved
add momentum (beta=0.9) at equal step-size 0.005SGD · Synthetic logistic regressionMSD: mu=mu_m=0.005, beta=0.9Mixed
add momentum beta=0.9 with mu_m=0.01 vs SGD mu=0.1SGD · Adult dataset, logistic regressiontest accuracyNo clear effect
add momentum beta=0.9 with mu_m=0.01 vs SGD mu=0.1SGD · CIFAR-10 (6-layer FC net)training accuracy: 90% -> 90%No clear effect
add momentum beta=0.9 with mu_m=0.001 vs SGD mu=0.01SGD · CIFAR-10 (4-layer CNN)training accuracy: 94%No clear effect
vary beta over 0.5, 0.6, 0.7, 0.8 instead of 0.9 (ablation)Momentum vs standard LMS · Synthetic LMS, 10-dimequivalence behaviourNo clear effect

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