Research map / Optimization / Distributed and compressed
Distributed Second Order Methods with Fast Rates and Compressed Communication
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 NEWTON-LEARN (NL1) instead of Newton's method | Distributed logistic regression · LIBSVM datasets and artificial dataset | communication complexity | Helped |
| Decrease the random-sparsification parameter r (ablation) | NEWTON-LEARN (NL1, NL2) · LIBSVM datasets and artificial dataset | communication complexity | Helped |
| Set r=1 and p=1 (ablation) | NEWTON-LEARN (NL1) · Compression-parameter experiments | communication complexity | Helped |
| Set r=1 and p=1/20 (ablation) | NEWTON-LEARN (NL2) · Compression-parameter experiments | communication complexity | Helped |
| Use NEWTON-LEARN (NL1) instead of BFGS | Distributed logistic regression · LIBSVM datasets and artificial dataset | communication complexity | Helped |
| Use NEWTON-LEARN (NL1) instead of ADIANA | Distributed logistic regression · LIBSVM datasets and artificial dataset | communication complexity | Helped |
| Use NEWTON-LEARN (NL2) instead of ADIANA | Distributed logistic regression · w8a and other tested datasets | communication complexity | Mixed |
| Use NEWTON-LEARN (NL1) instead of DINGO | Distributed logistic regression · Tested datasets | communication complexity | Helped |
| Use NEWTON-LEARN (NL2) instead of DINGO | Distributed logistic regression · Tested datasets | communication complexity | Mixed |
| Use CUBIC-NEWTON-LEARN (CNL) instead of DCGD or DIANA | Distributed logistic regression · λ=10^-3 | communication complexity | Helped |
| Use CUBIC-NEWTON-LEARN (CNL) instead of DCGD or DIANA | Distributed logistic regression · λ=10^-4 and λ=10^-5 | communication complexity | Helped |
| Use Newton's method instead of NEWTON-LEARN (NL1, NL2) | Distributed logistic regression · Tested datasets | iteration complexity | Helped |
Closest papers by meaning
- Communication-Efficient Distributed Optimization with Quantized Preconditioners (ICML 2021 · similarity 0.92)
- Accelerated Methods with Compressed Communications for Distributed Optimization Problems Under Data Similarity (AAAI 2025 · similarity 0.90)
- Scalable Distributed Stochastic Optimization via Bidirectional Compression: Beyond Pessimistic Limits (NeurIPS 2026 · similarity 0.90)
- Skip the Hessian, Keep the Rates: Globalized Semismooth Newton with Lazy Hessian Updates (NeurIPS 2026 · similarity 0.89)
- Towards Faster Decentralized Stochastic Optimization with Communication Compression (ICLR 2025 · similarity 0.89)
- Near-Optimal Distributed Minimax Optimization under the Second-Order Similarity (NeurIPS 2024 · similarity 0.89)
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.