Research map / Optimization / Distributed and compressed
Newton Method over Networks is Fast up to the Statistical Precision
Amir Daneshmand, Gesualdo Scutari, Pavel Dvurechensky, Alexander Gasnikov
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 DiRegINA rather than accelerated first-order methods | Distributed ridge regression · Synthetic ridge regression with β/μ larger than √κ | communication rounds to reduce function residual: from two to five time faster | Improved |
| Use DiRegINA rather than SONATA-L | Distributed logistic regression · a4a dataset | function residual convergence | Improved |
| Use DiRegINA rather than SONATA-F | Distributed logistic regression · Synthetic data with less local samples and larger dimension | function residual convergence | Improved |
| Use DiRegINA rather than SONATA-F | Distributed ridge regression · Scaled mg dataset | function residual convergence | Improved |
| Use DiRegINA rather than NN-1 | Distributed ridge regression · Scaled mg dataset | function residual convergence | Improved |
Closest papers by meaning
- Communication-Efficient Distributed Optimization in Networks with Gradient Tracking and Variance Reduction (JMLR 2020 · similarity 0.89)
- GIANT: Globally Improved Approximate Newton Method for Distributed Optimization (NeurIPS 2018 · similarity 0.88)
- Distributed Second Order Methods with Fast Rates and Compressed Communication (ICML 2021 · similarity 0.88)
- On Convergence of Distributed Approximate Newton Methods: Globalization, Sharper Bounds and Beyond (JMLR 2020 · similarity 0.88)
- Decentralized Sparse Linear Regression via Gradient-Tracking (JMLR 2025 · similarity 0.87)
- Communication Efficient Distributed Newton Method over Unreliable Networks (AAAI 2024 · similarity 0.85)
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.
