Research map / Optimization / Distributed and compressed
Efficient Distributed Learning with Sparsity
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 iterative shifted ℓ1-regularized updates with communicated local gradients instead of one-shot averaging of debiased local estimates | EDSL · Distributed sparse linear regression | estimation error | Helped |
| Use one EDSL communication round rather than Avg-Debias | Distributed sparse learning · Simulated regression and classification | estimation error | No clear effect |
| Use EDSL for a few communication rounds rather than stop after one-shot Avg-Debias | Distributed sparse learning · Simulated regression and classification | estimation error | Helped |
| Use EDSL rather than distributed proximal gradient descent | Distributed sparse learning · Ill-conditioned simulated problems | estimation error | Helped |
| Use EDSL rather than Avg-Debias | Distributed sparse learning · Real-world datasets | normalized MSE or misclassification error | Helped |
| Use EDSL rather than the centralized approach | Distributed sparse learning · Some real-world datasets | predictive performance | Mixed |
Closest papers by meaning
- Distributed Machine Learning with Sparse Heterogeneous Data (NeurIPS 2021 · similarity 0.90)
- Feature-distributed sparse regression: a screen-and-clean approach (NeurIPS 2016 · similarity 0.90)
- Gradient Sparsification for Communication-Efficient Distributed Optimization (NeurIPS 2018 · similarity 0.90)
- Distributed Sparse Regression via Penalization (NeurIPS 2024 · similarity 0.89)
- Communication-Efficient Gradient Descent-Accent Methods for Distributed Variational Inequalities: Unified Analysis and Local Updates (ICLR 2024 · 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.