Research map / Federated learning and privacy / Communication-efficient FL
Lower Bounds and Optimal Algorithms for Personalized 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 AL2SGD+ instead of L2SGD+ | Personalized FL with finite-sum local objectives · LIBSVM logistic regression; communication rounds | communication rounds to convergence | Helped |
| Use IAPGD+Katyusha instead of L2SGD+ | Personalized FL with finite-sum local objectives · LIBSVM logistic regression; communication rounds | communication rounds to convergence | Helped |
| Use AL2SGD+ instead of IAPGD+Katyusha | Personalized FL with finite-sum local objectives · LIBSVM logistic regression; local computation | local computation | Helped |
| Use IAPGD+Katyusha instead of L2SGD+ | Personalized FL with finite-sum local objectives · LIBSVM logistic regression; local computation | local computation | Hurt |
| Split client data heterogeneously rather than homogeneously | Personalized FL algorithms in the LIBSVM experiment · LIBSVM logistic regression; convergence | convergence | No clear effect |
| Use APGD1 when λ ≤ L=1 rather than APGD2 | Accelerated proximal-gradient methods · Synthetic quadratic objective; communication rounds | communication rounds | Helped |
| Use APGD2 when λ > L=1 rather than APGD1 | Accelerated proximal-gradient methods · Synthetic quadratic objective; communication rounds | communication rounds | Helped |
Closest papers by meaning
- On Sample Optimality in Personalized Collaborative and Federated Learning (NeurIPS 2022 · similarity 0.89)
- Understanding the Statistical Accuracy-Communication Trade-off in Personalized Federated Learning with Minimax Guarantees (ICML 2025 · similarity 0.89)
- Federated Learning under Arbitrary Communication Patterns (ICML 2021 · similarity 0.88)
- Solving a Class of Non-Convex Minimax Optimization in Federated Learning (NeurIPS 2023 · similarity 0.88)
- Personalized Federated Learning with First Order Model Optimization (ICLR 2021 · similarity 0.88)
- Bayesian Coreset Optimization for Personalized Federated Learning (ICLR 2024 · similarity 0.87)
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.