Research map / Optimization / Distributed and compressed
Delayed Algorithms for Distributed Stochastic Weakly Convex Optimization
Wenzhi Gao, Qi Deng
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 changed | Where | Measured | Effect |
|---|---|---|---|
| Use delayed prox-linear (DSPL) instead of delayed subgradient (DSGD) | delayed stochastic method · Robust phase retrieval, async MPI | function value gap / distance to optimum: 10−2 -> 10−6 | Improved |
| Increase number of workers | DSGD and DSPL · Robust phase retrieval, async MPI | wall-clock time for 400 epochs | Improved |
| Use DSPL instead of DSGD | delayed method · Robust phase retrieval, async MPI | wall-clock time | Hurt |
| Increase number of workers (few workers) | DSGD and DSPL · Robust phase retrieval, async MPI | convergence | No clear effect |
| Use DSPL instead of DSGD under simulated geometric/Poisson delay | delayed method · Robust phase retrieval, simulated delay | range of stepsize parameter alpha ensuring convergence | Improved |
| Increase simulated delay | DSGD and DSPL · Robust phase retrieval, simulated delay | iterations to converge | Hurt |
| Add safeguarding step to DSGD (ablation) | DSGD · Adversarial delay, zipcode and synthetic | function value | Improved |
| Add safeguarding step to DSPL (ablation) | DSPL · Adversarial delay | function value | Improved |
Closest papers by meaning
- Asynchronous Stochastic Optimization Robust to Arbitrary Delays (NeurIPS 2021 · similarity 0.89)
- Stochastic Weakly Convex Optimization beyond Lipschitz Continuity (ICML 2024 · similarity 0.88)
- Decentralized Convex Finite-Sum Optimization with Better Dependence on Condition Numbers (ICML 2024 · similarity 0.87)
- High-Probability Convergence for Composite and Distributed Stochastic Minimization and Variational Inequalities with Heavy-Tailed Noise (ICML 2024 · similarity 0.87)
- Faster Stochastic Optimization with Arbitrary Delays via Adaptive Asynchronous Mini-Batching (ICML 2025 · similarity 0.87)
- Single-Loop Stochastic Algorithms for Difference of Max-Structured Weakly Convex Functions (NeurIPS 2024 · similarity 0.87)
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.
