Research map / Optimization / Distributed and compressed
Local SGD with Periodic Averaging: Tighter Analysis and Adaptive Synchronization
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 changed | Where | Measured | Effect |
|---|---|---|---|
| Use LUPA-SGD instead of the cited prior periodic-averaging method | Local SGD · Non-convex optimization under PL condition | communication rounds: O(p^{1/2}T^{1/2}) -> O(p^{1/3}T^{1/3}) | Helped |
| Increase the local-update period during training (ablation) | ADA-LUPA-SGD · Epsilon | error rate | Helped |
| Increase SyncSGD mini-batch size from 512 to 1024 (ablation) | SyncSGD · Epsilon | convergence | Hurt |
Closest papers by meaning
- Communication-efficient SGD: From Local SGD to One-Shot Averaging (NeurIPS 2021 · similarity 0.92)
- Do We Need Asynchronous SGD? On the Near-Optimality of Synchronous Methods (NeurIPS 2026 · similarity 0.90)
- LOSCAR-SGD: Local SGD with Communication-Computation Overlap and Delay-Corrected Sparse Model Averaging (NeurIPS 2026 · similarity 0.89)
- Convergence of Distributed Adaptive Optimization with Local Updates (ICLR 2025 · similarity 0.89)
- SLowcalSGD : Slow Query Points Improve Local-SGD for Stochastic Convex Optimization (NeurIPS 2024 · similarity 0.89)
- What’s in a Smoothness Constant? Tight Rates for Local SGD with Bounded Second-order Heterogeneity (NeurIPS 2026 · similarity 0.88)
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.