Research map / Federated learning and privacy / Personalized and heterogeneous FL
FedDAG: Clustered Federated Learning via Global Data and Gradient Integration for Heterogeneous Environments
Anik Pramanik, Murat Kantarcioglu, Vincent Oria, Shantanu Sharma
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 global representation sharing with dual encoder (ablation) | FedDAG* · CIFAR-10 (20% label skew) | accuracy (%): 88.67 -> 90.76 | Improved |
| add global representation sharing with dual encoder (ablation) | FedDAG* · FMNIST (20% label skew) | accuracy (%): 92.75 -> 93.82 | Improved |
| add global representation sharing with dual encoder (ablation) | FedDAG* · SVHN (20% label skew) | accuracy (%): 91.87 -> 93.91 | Improved |
| add global representation sharing with dual encoder (ablation) | FedDAG* · CIFAR-100 (20% label skew) | accuracy (%): 70.37 -> 72.84 | Improved |
| add cross-cluster sharing to a dual encoder (ablation) | FedDAG-dagger · CIFAR-10 (20% label skew) | accuracy (%): 88.79 -> 90.76 | Improved |
| add cross-cluster sharing to a dual encoder (ablation) | FedDAG-dagger · SVHN (20% label skew) | accuracy (%): 91.95 -> 93.91 | Improved |
| add a second encoder without sharing to the single-encoder model (ablation) | FedDAG* · CIFAR-10 (20% label skew) | accuracy (%): 88.67 -> 88.79 | No clear effect |
| add second encoder without sharing to the single-encoder model (ablation) | FedDAG* · CIFAR-100 (20% label skew) | accuracy (%): 70.37 -> 70.28 | No clear effect |
| combine data and gradient similarity for clustering instead of PACFL data-only | FedDAG* · CIFAR-10 (20% label skew) | accuracy (%): 86.93 -> 88.67 | Improved |
| combine data and gradient similarity instead of IFCA gradient-only | FedDAG* · CIFAR-10 (20% label skew) | accuracy (%): 86.64 -> 88.67 | Improved |
| use FedDAG instead of best baseline CFL-GP under concept shift | FedDAG · CIFAR-10 (concept shift) | accuracy (%): 66.74 -> 69.90 | Improved |
| use FedDAG instead of IFCA under LDA skew | FedDAG · CIFAR-10 (LDA skew, alpha 0.25) | accuracy (%): 74.43 -> 77.84 | Improved |
Later papers that compared with this method
1 later accepted paper reports FedDAG as a baseline in their results tables; 1 of them report a better number on most of the results they share with it.
- Regret-Based Federated Causal Discovery with Unknown Interventions (ICML 2026 · better on 5 of 6 shared results)
Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.
Closest papers by meaning
- Enhancing Clustered Federated Learning: Integration of Strategies and Improved Methodologies (ICLR 2025 · similarity 0.90)
- Clustered Federated Learning via Gradient-based Partitioning (ICML 2024 · similarity 0.89)
- Interaction-Aware Gaussian Weighting for Clustered Federated Learning (ICML 2025 · similarity 0.88)
- FLUX: Efficient Descriptor-Driven Clustered Federated Learning under Arbitrary Distribution Shifts (NeurIPS 2025 · similarity 0.88)
- One-Shot Heterogeneous Federated Learning with Local Model-Guided Diffusion Models (ICML 2025 · similarity 0.88)
- FedGPS: Statistical Rectification Against Data Heterogeneity in Federated Learning (NeurIPS 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.
