Research map / Optimization / Nonconvex and smooth optimization
FedNest: Federated Bilevel, Minimax, and Compositional Optimization
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 FedNest instead of LFedNest | FedNest · MNIST hyper-representation, non-i.i.d. | test accuracy | Helped |
| Use FedNest instead of LFedNest | FedNest · MNIST hyper-representation, i.i.d. | test accuracy | No clear effect |
| Add SVRG to the inner optimizer (ablation) | FedNest · MNIST hyper-representation, especially non-i.i.d. | test accuracy | Helped |
| Add SVRG to the inner optimizer (ablation) | FedNest · Imbalanced MNIST loss-function tuning | test accuracy | Helped |
| Increase the outer optimizer local epoch $\tau$ (ablation) | FedNest · MNIST hyper-representation | test performance | Helped |
| Compute hypergradients locally rather than using FedNest's global hypergradient procedure (ablation) | FedNest · MNIST hyper-representation, heterogeneous clients | test accuracy | Hurt |
| Use alternating updates rather than simultaneous FedAvg saddle-point updates | Federated minimax optimization · Federated minimax synthetic problem | convergence | Helped |
| Control client drift with FedNest's global variance-reduced updates rather than local hypergradient updates (ablation) | FedNest · Federated minimax synthetic problem | convergence | Helped |
| Use FedNest instead of non-federated bilevel optimization | FedNest · Imbalanced MNIST loss-function tuning | test accuracy | No clear effect |
Closest papers by meaning
- Zeroth-Order Methods for Nondifferentiable, Nonconvex, and Hierarchical Federated Optimization (NeurIPS 2023 · similarity 0.90)
- A Doubly Recursive Stochastic Compositional Gradient Descent Method for Federated Multi-Level Compositional Optimization (ICML 2024 · similarity 0.88)
- First-Order Federated Bilevel Learning (AAAI 2025 · similarity 0.88)
- A Federated Stochastic Multi-level Compositional Minimax Algorithm for Deep AUC Maximization (ICML 2024 · similarity 0.87)
- Federated Optimization with Doubly Regularized Drift Correction (ICML 2024 · similarity 0.87)
- FedQS: Optimizing Gradient and Model Aggregation for Semi-Asynchronous Federated Learning (NeurIPS 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.