Research map / Federated learning and privacy / Personalized and heterogeneous FL
FedPall: Prototype-based Adversarial and Collaborative Learning for Federated Learning with Feature Drift
Yong Zhang, Feng Liang, Guanghu Yuan, Min Yang, Chengming Li, Xiping Hu
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 prototype mixing plus Bernoulli masking before upload (ablation) | FedPall · Office-10 (privacy study) | top-1 accuracy: 65.76 -> 67.55 | Improved |
| add prototype mixing plus Bernoulli masking before upload (ablation) | FedPall · Office-10 (privacy study) | mutual information (DEMINE): 2.96 -> 2.10 | Improved |
| add Bernoulli masking on top of prototype mixture (ablation) | FedPall · Office-10 (privacy study) | top-1 accuracy: 64.46 -> 67.55 | Improved |
| add prototype mixture on top of Bernoulli masking (ablation) | FedPall · Office-10 (privacy study) | top-1 accuracy: 63.70 -> 67.55 | Improved |
| add KL and InfoNCE losses to CE loss (ablation) | FedPall · Office-10 | top-1 accuracy: nearly 4 percentage points | Improved |
| add InfoNCE loss to CE loss alone (ablation) | FedPall · Office-10 | top-1 accuracy | Hurt |
| add KL loss to CE loss alone (ablation) | FedPall · Office-10 | top-1 accuracy | Improved |
| add KL loss to CE loss alone (ablation) | FedPall · PACS | top-1 accuracy | Hurt |
| add global classifier trained on mixed features (ablation) | FedPall · Office-10 (Caltech) | top-1 accuracy: 49.33% on Caltech, a 15.55% improvement | Improved |
| add global classifier trained on mixed features (ablation) | FedPall · Digits (MNIST-M) | top-1 accuracy: 3.61% higher | Improved |
| add global classifier trained on mixed features (ablation) | FedPall · PACS | top-1 accuracy: gains of up to 3% | Improved |
| add global classifier trained on mixed features (ablation) | FedPall · Digits | top-1 accuracy | Mixed |
Later papers that compared with this method
2 later accepted papers report FedPall as a baseline in their results tables; 2 of them report a better number on most of the results they share with it.
- Prototype-guided Bilateral Alignment Multimodal Federated Learning (ICML 2026 · better on 3 of 3 shared results)
- Bayesian Evidence-Driven Prototype Evolution for Federated Domain Adaptation (ICLR 2026 · better on 2 of 2 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
- Taming Cross-Domain Representation Variance in Federated Prototype Learning with Heterogeneous Data Domains (NeurIPS 2024 · similarity 0.89)
- Tackling Feature-Classifier Mismatch in Federated Learning via Prompt-Driven Feature Transformation (NeurIPS 2025 · similarity 0.89)
- A Simple Data Augmentation for Feature Distribution Skewed Federated Learning (CVPR 2025 · similarity 0.88)
- FedImpro: Measuring and Improving Client Update in Federated Learning (ICLR 2024 · similarity 0.87)
- Personalized Federated Learning via Feature Distribution Adaptation (NeurIPS 2024 · similarity 0.87)
- Recurrent Early Exits for Federated Learning with Heterogeneous Clients (ICML 2024 · 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.
