Research map / Federated learning and privacy / Personalized and heterogeneous FL
Bayesian Nonparametric Federated Learning of Neural Networks
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 |
|---|---|---|---|
| Aggregate independently trained local networks with PFNM instead of using local models | PFNM · MNIST and CIFAR-10; homogeneous and heterogeneous partitions | test accuracy | Helped |
| Use PFNM instead of federated averaging | MNIST and CIFAR-10; one- and two-layer networks · MNIST and CIFAR-10; homogeneous and heterogeneous partitions | test accuracy | Helped |
| Use PFNM instead of an ensemble | MNIST and CIFAR-10; single communication · MNIST and CIFAR-10; homogeneous and heterogeneous partitions | test accuracy | Mixed |
| Allow additional communication rounds while applying PFNM | PFNM · MNIST and CIFAR-10; homogeneous and heterogeneous partitions | test accuracy | Helped |
| Use PFNM instead of federated averaging or downpour SGD | PFNM with limited communication · MNIST and CIFAR-10; homogeneous and heterogeneous partitions | communication rounds to reach a given performance | Helped |
| Allow additional communication rounds while applying PFNM | PFNM · MNIST and CIFAR-10; homogeneous and heterogeneous partitions | global model size | Helped |
| Increase the number of batches, reducing data per local model | PFNM and local networks · MNIST and CIFAR-10 | test accuracy | Mixed |
| Vary the prior variance and mass parameter (ablation) | PFNM · MNIST and CIFAR-10; single hidden layer, 25 batches | training accuracy | No clear effect |
| Vary the local-weight variance (ablation) | PFNM · Homogeneous partitioning; single hidden layer | training accuracy | No clear effect |
| Vary the local-weight variance (ablation) | PFNM · Heterogeneous partitioning | training accuracy | Mixed |
Closest papers by meaning
- A Bayesian Approach for Personalized Federated Learning in Heterogeneous Settings (NeurIPS 2024 · similarity 0.87)
- Federated Generalised Variational Inference: A Robust Probabilistic Federated Learning Framework (ICML 2025 · similarity 0.87)
- Adaptive Federated Learning with Auto-Tuned Clients (ICLR 2024 · similarity 0.87)
- Calibrated One Round Federated Learning with Bayesian Inference in the Predictive Space (AAAI 2024 · similarity 0.87)
- Improving Generalization in Federated Learning with Model-Data Mutual Information Regularization: A Posterior Inference Approach (NeurIPS 2024 · similarity 0.86)
- FedBEns: One-Shot Federated Learning based on Bayesian Ensemble (ICML 2025 · similarity 0.86)
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.