Research map / Optimization / Distributed and compressed
Gradient Coding: Avoiding Stragglers in Distributed Learning
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 |
|---|---|---|---|
| Adding gradient coding (FracRep/CycRep full-gradient recovery with s stragglers) instead of ignoring stragglers | logistic regression on Amazon Employee Access · Amazon EC2 Employee Access, n=10,20,30 t2.micro | generalization AUC | Helped |
| Replicating each partition s+1 times with coded gradient combinations (vs. naive uncoded division) | gradient descent on EC2 artificial dataset · Amazon EC2 artificial dataset, n=12, s=1 and s=2 | running time | Helped |
| Ignoring stragglers (updating after n-s workers) instead of waiting for all workers | gradient descent on EC2 · Amazon EC2 Employee Access | running time per iteration | Helped |
| Partial straggler two-stage scheme (naive plus coded partitions) vs full straggler schemes | gradient coding · Amazon EC2 artificial dataset, n=12 | running time | No clear effect |
Closest papers by meaning
- Approximate Gradient Coding for Distributed Learning with Heterogeneous Stragglers (NeurIPS 2025 · similarity 0.94)
- Leveraging partial stragglers within gradient coding (NeurIPS 2024 · similarity 0.93)
- Communication-Computation Efficient Gradient Coding (ICML 2018 · similarity 0.92)
- Gradient Coding from Cyclic MDS Codes and Expander Graphs (ICML 2018 · similarity 0.91)
- Rescaled Asynchronous SGD: Optimal Distributed Optimization under Data and System Heterogeneity (NeurIPS 2026 · similarity 0.87)
- Accelerated Methods with Compressed Communications for Distributed Optimization Problems Under Data Similarity (AAAI 2025 · similarity 0.86)
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.