Research map / Federated learning and privacy / Personalized and heterogeneous FL
Federated Representation Angle Learning
Liping Yi, Han Yu, Gang Wang, Xiaoguang Liu, Xiaoxiao Li
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 adaptive diagonal sparsification (ablation) | FedRAL · CIFAR-10 (pathological) | average accuracy (%): 91.64 -> 93.60 | Improved |
| add adaptive diagonal sparsification (ablation) | FedRAL · CIFAR-100 (pathological) | average accuracy (%): 56.58 -> 58.70 | Improved |
| add adaptive diagonal sparsification (ablation) | FedRAL · CIFAR-10 (practical) | average accuracy (%): 41.37 -> 42.97 | Improved |
| add adaptive diagonal sparsification (ablation) | FedRAL · CIFAR-100 (practical) | average accuracy (%): 11.58 -> 11.70 | No clear effect |
| use ordinate-wise weighted aggregation instead of naive averaging (ablation) | FedRAL · CIFAR-10 (pathological) | average accuracy (%): 93.47 -> 93.60 | No clear effect |
| use ordinate-wise weighted aggregation instead of naive averaging (ablation) | FedRAL · CIFAR-100 (pathological) | average accuracy (%): 53.94 -> 58.70 | Improved |
| use ordinate-wise weighted aggregation instead of naive averaging (ablation) | FedRAL · CIFAR-10 (practical) | average accuracy (%): 40.52 -> 42.97 | Improved |
| use ordinate-wise weighted aggregation instead of naive averaging (ablation) | FedRAL · CIFAR-100 (practical) | average accuracy (%): 9.24 -> 11.70 | Improved |
| use FedRAL instead of FedTGP | FedRAL · CIFAR-100 (practical non-IID) | average test accuracy (%): 9.37 -> 11.70 | Improved |
| use FedRAL instead of FedTGP | FedRAL · CIFAR-100 (pathological non-IID) | average test accuracy (%): 53.20 -> 58.70 | Improved |
| use FedRAL instead of FedTGP | FedRAL · CIFAR-100 (communication rounds) | rounds to target accuracy: 354 -> 144 | Improved |
| use FedRAL instead of FedTGP under high model heterogeneity | FedRAL · CIFAR-100 (100 clients) | average accuracy (%): 48.69 -> 50.79 | Improved |
Closest papers by meaning
- Federated Model Heterogeneous Matryoshka Representation Learning (NeurIPS 2024 · similarity 0.93)
- FedQS: Optimizing Gradient and Model Aggregation for Semi-Asynchronous Federated Learning (NeurIPS 2025 · similarity 0.87)
- Federated Representation Learning in the Under-Parameterized Regime (ICML 2024 · similarity 0.87)
- FedSA: A Unified Representation Learning via Semantic Anchors for Prototype-based Federated Learning (AAAI 2025 · similarity 0.87)
- Adaptive Latent-Space Constraints in Personalized Federated Learning (NeurIPS 2025 · similarity 0.87)
- Subspace Constraint and Contribution Estimation for Heterogeneous Federated Learning (CVPR 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.
