Research map / Federated learning and privacy / Communication-efficient FL
Fair Resource Allocation in 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 |
|---|---|---|---|
| Increase q from 0 to 1 to emphasize devices with higher loss (ablation) | q-FFL · Synthetic | test accuracy variance: 724 ± 72 -> 472 ± 14 | Helped |
| Increase q from 0 to 5 to emphasize devices with higher loss (ablation) | q-FFL · Vehicle | test accuracy variance: 291 ± 18 -> 48 ± 5 | Helped |
| Increase q from 0 to 1 to emphasize devices with higher loss (ablation) | q-FFL · Sent140 | test accuracy variance: 697 ± 132 -> 509 ± 30 | Helped |
| Increase q from 0 to .001 to emphasize devices with higher loss (ablation) | q-FFL · Shakespeare | test accuracy variance: 82 ± 41 -> 54 ± 27 | Helped |
| Increase q from 0 to 1 to emphasize devices with higher loss (ablation) | q-FFL · Synthetic | worst 10% test accuracy: 18.8 ± 5.0 -> 31.1 ± 1.8 | Helped |
| Increase q from 0 to 5 to emphasize devices with higher loss (ablation) | q-FFL · Vehicle | worst 10% test accuracy: 43.0 ± 1.0 -> 69.9 ± .6 | Helped |
| Increase q from 0 to 1 to emphasize devices with higher loss (ablation) | q-FFL · Sent140 | worst 10% test accuracy: 15.9 ± 4.9 -> 23.0 ± 1.4 | Helped |
| Increase q from 0 to .001 to emphasize devices with higher loss (ablation) | q-FFL · Shakespeare | worst 10% test accuracy: 39.7 ± 2.8 -> 42.1 ± 2.1 | Helped |
| Use q1>0 rather than q=0 (ablation) | q-FFL · Adult | PhD accuracy: 69.9 ± .4 -> 74.1 ± .6 | Helped |
| Use q1>0 rather than q=0 (ablation) | q-FFL · Fashion MNIST | shirt accuracy: 66.0 ± .7 -> 74.2 ± .3 | Helped |
| Use local updating instead of non-local gradient updates | q-FedAvg versus q-FedSGD · Most evaluated federated datasets | communication-round convergence speed | Helped |
| Use local updating instead of non-local gradient updates | q-FedAvg versus q-FedSGD · Synthetic | communication-round convergence speed | Hurt |
Closest papers by meaning
- FedLF: Layer-Wise Fair Federated Learning (AAAI 2024 · similarity 0.89)
- Pursuing Overall Welfare in Federated Learning through Sequential Decision Making (ICML 2024 · similarity 0.87)
- FairTrade: Achieving Pareto-Optimal Trade-Offs between Balanced Accuracy and Fairness in Federated Learning (AAAI 2024 · similarity 0.87)
- FedFACT: A Provable Framework for Controllable Group-Fairness Calibration in Federated Learning (NeurIPS 2025 · similarity 0.87)
- Fed-QSSL: A Framework for Personalized Federated Learning under Bitwidth and Data Heterogeneity (AAAI 2024 · similarity 0.86)
- Demystifying Local & Global Fairness Trade-offs in Federated Learning Using Partial Information Decomposition (ICLR 2024 · 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.