Research map / Federated learning and privacy / Personalized and heterogeneous FL
Federated Multi-Task Learning under a Mixture of Distributions
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 |
|---|---|---|---|
| mixture-of-distributions personalization (FedEM) vs best baseline | federated learning · CIFAR10 | average accuracy: 82.3 -> 84.3 | Helped |
| FedEM vs pFedMe | personalized FL · Synthetic | average accuracy: 69.2 -> 74.7 | Helped |
| FedEM vs FedAvg | federated learning · Shakespeare | average accuracy: 46.7 -> 46.7 | No clear effect |
| FedEM vs pFedMe | personalized FL · CIFAR10 | bottom decile accuracy: 73.6 -> 78.1 | Helped |
| FedEM personalization for unseen clients vs FedAvg+ | unseen clients · CIFAR100 | average accuracy: 36.5 -> 47.5 | Helped |
| FedEM vs FedAvg+ under low client participation | client sampling · CIFAR10 | accuracy | Helped |
| increase mixture components M beyond 3 (ablation) | FedEM · CIFAR10 | accuracy | No clear effect |
| limit FedEM to one third of rounds (ablation) | FedEM · multiple datasets | accuracy | Mixed |
Closest papers by meaning
- FedHCA2: Towards Hetero-Client Federated Multi-Task Learning (CVPR 2024 · similarity 0.89)
- Towards the Theory of Unsupervised Federated Learning: Non-asymptotic Analysis of Federated EM Algorithms (ICML 2024 · similarity 0.88)
- A Bayesian Approach for Personalized Federated Learning in Heterogeneous Settings (NeurIPS 2024 · similarity 0.88)
- Adaptive Latent-Space Constraints in Personalized Federated Learning (NeurIPS 2025 · similarity 0.88)
- FedSPU: Personalized Federated Learning for Resource-Constrained Devices with Stochastic Parameter Update (AAAI 2025 · similarity 0.88)
- Towards Diverse Device Heterogeneous Federated Learning via Task Arithmetic Knowledge Integration (NeurIPS 2024 · 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.