Federated Adapter on Foundation Models: An Out-Of-Distribution Approach

Fuente: arXiv
Guardado en:
Detalles Bibliográficos
Autores principales: Yang, Yiyuan, Long, Guodong, Zhou, Tianyi, Lu, Qinghua, Ye, Shanshan, Jiang, Jing
Formato: Preprint
Publicado: 2025
Materias:
Acceso en línea:
Etiquetas: Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
_version_ 1866912357710561280
author Yang, Yiyuan
Long, Guodong
Zhou, Tianyi
Lu, Qinghua
Ye, Shanshan
Jiang, Jing
author_facet Yang, Yiyuan
Long, Guodong
Zhou, Tianyi
Lu, Qinghua
Ye, Shanshan
Jiang, Jing
contents As foundation models gain prominence, Federated Foundation Models (FedFM) have emerged as a privacy-preserving approach to collaboratively fine-tune models in federated learning (FL) frameworks using distributed datasets across clients. A key challenge for FedFM, given the versatile nature of foundation models, is addressing out-of-distribution (OOD) generalization, where unseen tasks or clients may exhibit distribution shifts leading to suboptimal performance. Although numerous studies have explored OOD generalization in conventional FL, these methods are inadequate for FedFM due to the challenges posed by large parameter scales and increased data heterogeneity. To address these, we propose FedOA, which employs adapter-based parameter-efficient fine-tuning methods for efficacy and introduces personalized adapters with feature distance-based regularization to align distributions and guarantee OOD generalization for each client. Theoretically, we demonstrate that the conventional aggregated global model in FedFM inherently retains OOD generalization capabilities, and our proposed method enhances the personalized model's OOD generalization through regularization informed by the global model, with proven convergence under general non-convex settings. Empirically, the effectiveness of the proposed method is validated on benchmark datasets across various NLP tasks.
format Preprint
id arxiv_https___arxiv_org_abs_2505_01075
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Federated Adapter on Foundation Models: An Out-Of-Distribution Approach
Yang, Yiyuan
Long, Guodong
Zhou, Tianyi
Lu, Qinghua
Ye, Shanshan
Jiang, Jing
Machine Learning
As foundation models gain prominence, Federated Foundation Models (FedFM) have emerged as a privacy-preserving approach to collaboratively fine-tune models in federated learning (FL) frameworks using distributed datasets across clients. A key challenge for FedFM, given the versatile nature of foundation models, is addressing out-of-distribution (OOD) generalization, where unseen tasks or clients may exhibit distribution shifts leading to suboptimal performance. Although numerous studies have explored OOD generalization in conventional FL, these methods are inadequate for FedFM due to the challenges posed by large parameter scales and increased data heterogeneity. To address these, we propose FedOA, which employs adapter-based parameter-efficient fine-tuning methods for efficacy and introduces personalized adapters with feature distance-based regularization to align distributions and guarantee OOD generalization for each client. Theoretically, we demonstrate that the conventional aggregated global model in FedFM inherently retains OOD generalization capabilities, and our proposed method enhances the personalized model's OOD generalization through regularization informed by the global model, with proven convergence under general non-convex settings. Empirically, the effectiveness of the proposed method is validated on benchmark datasets across various NLP tasks.
title Federated Adapter on Foundation Models: An Out-Of-Distribution Approach
topic Machine Learning
url https://arxiv.org/abs/2505.01075