Research map / Federated learning and privacy / Personalized and heterogeneous FL
Structured Federated Learning through Clustered Additive Modeling
Jie Ma, Tianyi Zhou, Guodong Long, Jing Jiang, Chengqi Zhang
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (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 |
|---|---|---|---|
| add CAM to IFCA | IFCA · CIFAR-10 cluster-wise Dirichlet, K=10 | accuracy: 34.84 -> 70.90 | Improved |
| add CAM to IFCA | IFCA · Fashion-MNIST cluster-wise n-class (3,2), K=10 | macro-F1: 66.22 -> 82.98 | Improved |
| add CAM to FeSEM | FeSEM · CIFAR-10 cluster-wise Dirichlet, K=10 | accuracy: 66.89 -> 78.45 | Improved |
| add CAM to FeSEM | FeSEM · Fashion-MNIST cluster-wise n-class (3,2), K=10 | macro-F1: 84.43 -> 92.43 | Improved |
| add CAM to FeSEM | FeSEM · CIFAR-10 client-wise Dirichlet 0.1, K=10 | macro-F1: 31.69 -> 59.19 | Improved |
| add CAM to IFCA | IFCA · CIFAR-10 client-wise Dirichlet 0.1, K=10 | accuracy: 47.62 -> 72.54 | Improved |
| add CAM to IFCA | IFCA · Fashion-MNIST client-wise n-class (2), K=5 | macro-F1: 72.30 -> 70.24 | No clear effect |
| ensemble FedAvg K=10 times (FedAvg+) | FedAvg · Fashion-MNIST cluster-wise Dirichlet, K=10 | accuracy: 86.08 -> 89.42 | No clear effect |
| increase number of clusters K from 5 to 10 with CAM | FeSEM-CAM · CIFAR-10 cluster-wise Dirichlet | accuracy: 64.35 -> 78.45 | Improved |
| increase number of clusters K from 5 to 10 with CAM | IFCA-CAM · CIFAR-10 cluster-wise Dirichlet | accuracy: 58.13 -> 70.90 | Improved |
| increase K from 5 to 10 without CAM | IFCA · Fashion-MNIST cluster-wise Dirichlet | accuracy: 84.60 -> 82.10 | Hurt |
Closest papers by meaning
- Enhancing Clustered Federated Learning: Integration of Strategies and Improved Methodologies (ICLR 2025 · similarity 0.91)
- Clustered Federated Learning via Gradient-based Partitioning (ICML 2024 · similarity 0.90)
- FLUX: Efficient Descriptor-Driven Clustered Federated Learning under Arbitrary Distribution Shifts (NeurIPS 2025 · similarity 0.89)
- Cluster Based Heterogeneous Federated Foundation Model Adaptation and Fine-Tuning (AAAI 2025 · similarity 0.89)
- Interaction-Aware Gaussian Weighting for Clustered Federated Learning (ICML 2025 · similarity 0.89)
- One-Shot Heterogeneous Federated Learning with Local Model-Guided Diffusion Models (ICML 2025 · similarity 0.88)
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.
