AI Paper Atlas

Research map / Optimization / Nonconvex and smooth optimization

Decentralized Learning: Theoretical Optimality and Practical Improvements

Yucheng Lu, Christopher De Sa

JMLR 2023 · Optimization › Nonconvex and smooth optimization · approach: Proximal finite-sum methods

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 DeTAG instead of D-PSGD on increasingly unshuffled (decentralized) dataLeNet 8 workers ring · CIFAR10 (X% shuffled)training loss convergenceImproved
use DeTAG instead of D2 and DSGT on unshuffled dataLeNet 8 workers ring · CIFAR10 (X% shuffled)training loss convergenceImproved
use DeTAG on fully shuffled dataLeNet 8 workers ring · CIFAR10 (100% shuffled)training loss convergenceNo clear effect
use DeTAG instead of baselines as spectral gap shrinksResnet20 16 workers ring · CIFAR100 (varying spectral gap)training loss convergenceImproved
use DeTAG* instead of D-PSGD, D2, DSGTResnet18 32-GPU ring · ImageNettraining loss and validation accuracy per roundImproved
add multi-round gossip (R-1 extra rounds) to larger mini-batch (ablation)DeTAG* vs naive R-times batch · ImageNet (R=2,4,8)validation 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