Research map / Federated learning and privacy / Communication-efficient FL
Achieving Linear Speedup in Non-IID Federated Bilevel 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 FedMBO with partial client participation instead of FedNest | Non-IID federated bilevel optimization · Non-IID federated bilevel optimization | sample complexity: \mathcal{O}(\epsilon^{-2}) -> \mathcal{O}(\epsilon^{-2} n^{-1}) | Helped |
| Use PHE + MinibatchSGD instead of the compared combinations of PHE, IHGP, MinibatchSGD, FedSVRG, and FedAvg | Federated bilevel hyper-representation · MNIST, heterogeneous clients | loss vs communication rounds | Helped |
| Use MinibatchSGD instead of FedAvg for lower-level updates (ablation) | Federated bilevel hyper-representation · MNIST, heterogeneous clients | lower-level loss | Helped |
| Use MinibatchSGD instead of FedSVRG for lower-level updates (ablation) | Federated bilevel hyper-representation · MNIST, heterogeneous clients | lower-level loss | No clear effect |
| Use PHE + MinibatchSGD instead of the compared alternatives (ablation) | Federated bilevel hyper-representation · MNIST, heterogeneous clients | test accuracy | Helped |
| Increase client data diversity from the 1-digit case toward the 10-digit case (ablation) | FedMBO · MNIST, varying client heterogeneity | test accuracy | Helped |
| Increase client sampling ratio (ablation) | FedMBO · MNIST, 2-digit client data | test accuracy | Helped |
| Increase the number of total clients (ablation) | FedMBO · MNIST, 2-digit client data and 10\% sampling ratio | test accuracy | Helped |
| Use LFedNest instead of FedNest (ablation) | Federated bilevel hyper-representation · Non-IID experiments | convergence | Hurt |
Closest papers by meaning
- First-Order Federated Bilevel Learning (AAAI 2025 · similarity 0.92)
- Exact and Linear Convergence for Federated Learning under Arbitrary Client Participation is Attainable (NeurIPS 2025 · similarity 0.87)
- Federated Learning under Periodic Client Participation and Heterogeneous Data: A New Communication-Efficient Algorithm and Analysis (NeurIPS 2024 · similarity 0.87)
- TurboSVM-FL: Boosting Federated Learning through SVM Aggregation for Lazy Clients (AAAI 2024 · similarity 0.87)
- FedNS: A Fast Sketching Newton-Type Algorithm for Federated Learning (AAAI 2024 · similarity 0.87)
- Methods with Local Steps and Random Reshuffling for Generally Smooth Non-Convex Federated Optimization (ICLR 2025 · 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.