Research map / Federated learning and privacy / Communication-efficient FL
FedNL: Making Newton-Type Methods Applicable to Federated 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 |
|---|---|---|---|
| Use learned and compressed local Hessian updates instead of repeatedly communicating full Hessians | FedNL · LibSVM logistic-regression experiments | communication complexity | Helped |
| Use backtracking line search rather than cubic regularization | FedNL global convergence · LibSVM logistic-regression experiments | communication complexity | Helped |
| Add cubic regularization instead of using line search | FedNL · LibSVM logistic-regression experiments | communication complexity | Hurt |
| Use Rank-1 Hessian compression rather than the other tested compressors (ablation) | FedNL · LibSVM logistic-regression experiments | communication complexity | Helped |
| Reduce the compression parameter for Rank-R, Top-K, or PowerSGD (ablation) | FedNL · LibSVM logistic-regression experiments | communication complexity | Helped |
| Use Hessian projection (Option 1) instead of error correction (Option 2) (ablation) | FedNL · LibSVM logistic-regression experiments | communication complexity | Helped |
| Replace Rank-R compression with Top-K or PowerSGD compression (ablation) | FedNL · LibSVM logistic-regression experiments | communication complexity | Mixed |
| Use FedNL-CR instead of first-order methods | FedNL global convergence · LibSVM logistic-regression experiments | communication complexity | Mixed |
Closest papers by meaning
- FedNS: A Fast Sketching Newton-Type Algorithm for Federated Learning (AAAI 2024 · similarity 0.90)
- FedQS: Optimizing Gradient and Model Aggregation for Semi-Asynchronous Federated Learning (NeurIPS 2025 · similarity 0.87)
- Federated Optimization with Doubly Regularized Drift Correction (ICML 2024 · similarity 0.86)
- Achieving Lossless Gradient Sparsification via Mapping to Alternative Space in Federated Learning (ICML 2024 · similarity 0.86)
- FedSMU: Communication-Efficient and Generalization-Enhanced Federated Learning through Symbolic Model Updates (ICML 2025 · similarity 0.86)
- SparsyFed: Sparse Adaptive Federated Learning (ICLR 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.