Research map / Optimization / Distributed and compressed
Asynchronous Distributed Semi-Stochastic Gradient Optimization
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 variance-reduced gradient in DPG (vr-dpg vs dpg) (ablation) | DPG · Mnist8m and DNA | objective vs wall-clock time | Helped |
| apply delayed gradient to current iterate v^t as well (distr-vr-sgd vs vr-dpg style update) (ablation) | distr-vr-sgd · Mnist8m and DNA | objective vs wall-clock time | Helped |
| set theta from nonzero to 0 (distr-svrg) (ablation) | distr-vr-sgd · Mnist8m and DNA | convergence speed: learning rate $10^{-3}$ -> $10^{-6}$ | Hurt |
| increase workers from 16 to 128 | distr-vr-sgd · Mnist8m | time to target objective: about 3 times faster | Helped |
| increase workers from 16 to 128 | distr-vr-sgd · DNA | time to target objective: about 6 times | Helped |
| increase delay bound tau from 10 to 200 | distr-vr-sgd · Mnist8m/DNA, 128 workers | communication cost | Helped |
Closest papers by meaning
- Sharper Convergence Guarantees for Asynchronous SGD for Distributed and Federated Learning (NeurIPS 2022 · similarity 0.93)
- Asynchronous Decentralized Parallel Stochastic Gradient Descent (ICML 2018 · similarity 0.92)
- Distributed Stochastic Optimization via Adaptive SGD (NeurIPS 2018 · similarity 0.91)
- Rescaled Asynchronous SGD: Optimal Distributed Optimization under Data and System Heterogeneity (NeurIPS 2026 · similarity 0.91)
- Asynchronous Mini-Batch Gradient Descent with Variance Reduction for Non-Convex Optimization (AAAI 2017 · similarity 0.91)
- Ringmaster ASGD: The First Asynchronous SGD with Optimal Time Complexity (ICML 2025 · similarity 0.90)
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.