AI Paper Atlas

Research map / Optimization / Nonconvex and smooth optimization

A Decentralized Proximal Gradient Tracking Algorithm for Composite Optimization on Riemannian Manifolds

Lei Wang, Le Bao, Xin Liu

JMLR 2025 · Optimization › Nonconvex and smooth optimization · approach: Saddle-point escape algorithms · 4 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 DR-ProxGT instead of DRSM and THANOSdecentralized CISE on MNIST · MNIST CISE (d=16 agents, p=10, mu=0.001; ring, grid, Erdos-Renyi)stationarity violation and consensus error vs communication roundsImproved
use DR-ProxGT instead of DRSMsparse PCA · Global Health Estimates, lambda = 0.05function value: -7.52 -> -7.76Improved
use DR-ProxGT instead of THANOSsparse PCA · Global Health Estimates, lambda = 0.05function value: -7.65 -> -7.76Improved
use DR-ProxGT instead of DRSMsparse PCA · Global Health Estimates, lambda = 0.1sparsity level: 30.53% -> 50.53%Improved
use DR-ProxGT instead of THANOSsparse PCA · Global Health Estimates, lambda = 0.15sparsity level: 34.74% -> 64.21%Improved
use DR-ProxGT instead of DRSMsparse PCA · Global Health Estimates, lambda = 0.05consensus error: 2.35e-04 -> 2.26e-04No clear effect
use DR-ProxGT instead of THANOSsparse PCA · Global Health Estimates, lambda = 0.05consensus error: 4.96e-04 -> 2.26e-04Improved
use DR-ProxGT instead of DRSMsparse PCA · Global Health Estimates, lambda = 0.15consensus error: 2.51e-04 -> 4.12e-05Improved

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