Federated Domain Generalization with Data-free On-server Matching Gradient
Fuente:
arXiv
Saved in:
| Main Authors: | , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2025
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866912393980805120 |
|---|---|
| author | Nguyen, Trong-Binh Nguyen, Minh-Duong Park, Jinsun Pham, Quoc-Viet Hwang, Won Joo |
| author_facet | Nguyen, Trong-Binh Nguyen, Minh-Duong Park, Jinsun Pham, Quoc-Viet Hwang, Won Joo |
| contents | Domain Generalization (DG) aims to learn from multiple known source domains a model that can generalize well to unknown target domains. One of the key approaches in DG is training an encoder which generates domain-invariant representations. However, this approach is not applicable in Federated Domain Generalization (FDG), where data from various domains are distributed across different clients. In this paper, we introduce a novel approach, dubbed Federated Learning via On-server Matching Gradient (FedOMG), which can \emph{efficiently leverage domain information from distributed domains}. Specifically, we utilize the local gradients as information about the distributed models to find an invariant gradient direction across all domains through gradient inner product maximization. The advantages are two-fold: 1) FedOMG can aggregate the characteristics of distributed models on the centralized server without incurring any additional communication cost, and 2) FedOMG is orthogonal to many existing FL/FDG methods, allowing for additional performance improvements by being seamlessly integrated with them. Extensive experimental evaluations on various settings to demonstrate the robustness of FedOMG compared to other FL/FDG baselines. Our method outperforms recent SOTA baselines on four FL benchmark datasets (MNIST, EMNIST, CIFAR-10, and CIFAR-100), and three FDG benchmark datasets (PACS, VLCS, and OfficeHome). |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2501_14653 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Federated Domain Generalization with Data-free On-server Matching Gradient Nguyen, Trong-Binh Nguyen, Minh-Duong Park, Jinsun Pham, Quoc-Viet Hwang, Won Joo Machine Learning Artificial Intelligence Distributed, Parallel, and Cluster Computing Multiagent Systems 68Q32, 68Q32 I.4.0; I.2.11 Domain Generalization (DG) aims to learn from multiple known source domains a model that can generalize well to unknown target domains. One of the key approaches in DG is training an encoder which generates domain-invariant representations. However, this approach is not applicable in Federated Domain Generalization (FDG), where data from various domains are distributed across different clients. In this paper, we introduce a novel approach, dubbed Federated Learning via On-server Matching Gradient (FedOMG), which can \emph{efficiently leverage domain information from distributed domains}. Specifically, we utilize the local gradients as information about the distributed models to find an invariant gradient direction across all domains through gradient inner product maximization. The advantages are two-fold: 1) FedOMG can aggregate the characteristics of distributed models on the centralized server without incurring any additional communication cost, and 2) FedOMG is orthogonal to many existing FL/FDG methods, allowing for additional performance improvements by being seamlessly integrated with them. Extensive experimental evaluations on various settings to demonstrate the robustness of FedOMG compared to other FL/FDG baselines. Our method outperforms recent SOTA baselines on four FL benchmark datasets (MNIST, EMNIST, CIFAR-10, and CIFAR-100), and three FDG benchmark datasets (PACS, VLCS, and OfficeHome). |
| title | Federated Domain Generalization with Data-free On-server Matching Gradient |
| topic | Machine Learning Artificial Intelligence Distributed, Parallel, and Cluster Computing Multiagent Systems 68Q32, 68Q32 I.4.0; I.2.11 |
| url | https://arxiv.org/abs/2501.14653 |