FedDAR: Federated Learning With Data-Quantity Aware Regularization for Heterogeneous Distributed Data
Federated learning (FL) has emerged as a promising approach for collaboratively training global models and classifiers without sharing private data. However, existing studies primarily focus on distinct methodologies for typical and personalized FL (tFL and pFL), representing a challenge in explorin...
Saved in:
| Main Authors: | , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
IEEE
2025-01-01
|
| Series: | IEEE Access |
| Subjects: | |
| Online Access: | https://ieeexplore.ieee.org/document/11091276/ |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1839605357495713792 |
|---|---|
| author | Youngjun Kwak Minyoung Jung |
| author_facet | Youngjun Kwak Minyoung Jung |
| author_sort | Youngjun Kwak |
| collection | DOAJ |
| description | Federated learning (FL) has emerged as a promising approach for collaboratively training global models and classifiers without sharing private data. However, existing studies primarily focus on distinct methodologies for typical and personalized FL (tFL and pFL), representing a challenge in exploring cross-applicable training methods. Moreover, previous approaches often rely on data and feature augmentation branches, overlooking data-quantity considerations, leading to suboptimal performance and inefficient communication costs, particularly in multi-class classification tasks. To address these challenges, we propose a novel add-on regularization technique for existing FL methods, named Data-quantity Aware Regularization (FedDAR), seamlessly integrating with existing tFL and pFL frameworks. This network-agnostic methodology reformulates the local training procedure by incorporating two crucial components: 1) enriched-feature augmentation, where features of the local model are coordinated with pre-initialized features to ensure unbiased-representations with efficient global communication rounds for unbalanced data distribution, and 2) data-quantity aware branch, which associates with local data size to improve the optimization of the local model using both supervised and self-supervised labels. We demonstrate significant performance improvements in tFL and pFL, achieving state-of-the-art results across MNIST, F-MNIST, CIFAR-10/100, and Tiny-ImageNet benchmarks. |
| format | Article |
| id | doaj-art-d5ed5f3b19d4472b9df8a4c36cc31c36 |
| institution | Matheson Library |
| issn | 2169-3536 |
| language | English |
| publishDate | 2025-01-01 |
| publisher | IEEE |
| record_format | Article |
| series | IEEE Access |
| spelling | doaj-art-d5ed5f3b19d4472b9df8a4c36cc31c362025-08-01T23:00:54ZengIEEEIEEE Access2169-35362025-01-011313320813321710.1109/ACCESS.2025.359183911091276FedDAR: Federated Learning With Data-Quantity Aware Regularization for Heterogeneous Distributed DataYoungjun Kwak0https://orcid.org/0009-0004-4805-8786Minyoung Jung1https://orcid.org/0009-0001-6995-5687Financial Tech Laboratory, KakaoBank Corporation, Seongnam-si, South KoreaKETI, Korea Electronics Technology Institute, Seongnam-si, South KoreaFederated learning (FL) has emerged as a promising approach for collaboratively training global models and classifiers without sharing private data. However, existing studies primarily focus on distinct methodologies for typical and personalized FL (tFL and pFL), representing a challenge in exploring cross-applicable training methods. Moreover, previous approaches often rely on data and feature augmentation branches, overlooking data-quantity considerations, leading to suboptimal performance and inefficient communication costs, particularly in multi-class classification tasks. To address these challenges, we propose a novel add-on regularization technique for existing FL methods, named Data-quantity Aware Regularization (FedDAR), seamlessly integrating with existing tFL and pFL frameworks. This network-agnostic methodology reformulates the local training procedure by incorporating two crucial components: 1) enriched-feature augmentation, where features of the local model are coordinated with pre-initialized features to ensure unbiased-representations with efficient global communication rounds for unbalanced data distribution, and 2) data-quantity aware branch, which associates with local data size to improve the optimization of the local model using both supervised and self-supervised labels. We demonstrate significant performance improvements in tFL and pFL, achieving state-of-the-art results across MNIST, F-MNIST, CIFAR-10/100, and Tiny-ImageNet benchmarks.https://ieeexplore.ieee.org/document/11091276/Federated learningregularizationclassificationaugmentationcommunication efficiencydata-quantity awareness |
| spellingShingle | Youngjun Kwak Minyoung Jung FedDAR: Federated Learning With Data-Quantity Aware Regularization for Heterogeneous Distributed Data IEEE Access Federated learning regularization classification augmentation communication efficiency data-quantity awareness |
| title | FedDAR: Federated Learning With Data-Quantity Aware Regularization for Heterogeneous Distributed Data |
| title_full | FedDAR: Federated Learning With Data-Quantity Aware Regularization for Heterogeneous Distributed Data |
| title_fullStr | FedDAR: Federated Learning With Data-Quantity Aware Regularization for Heterogeneous Distributed Data |
| title_full_unstemmed | FedDAR: Federated Learning With Data-Quantity Aware Regularization for Heterogeneous Distributed Data |
| title_short | FedDAR: Federated Learning With Data-Quantity Aware Regularization for Heterogeneous Distributed Data |
| title_sort | feddar federated learning with data quantity aware regularization for heterogeneous distributed data |
| topic | Federated learning regularization classification augmentation communication efficiency data-quantity awareness |
| url | https://ieeexplore.ieee.org/document/11091276/ |
| work_keys_str_mv | AT youngjunkwak feddarfederatedlearningwithdataquantityawareregularizationforheterogeneousdistributeddata AT minyoungjung feddarfederatedlearningwithdataquantityawareregularizationforheterogeneousdistributeddata |