Enhancing Federated Domain Adaptation with Multi-Domain Prototype-Based Federated Fine-Tuning

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Zhang, Jingyuan, Duan, Yiyang, Niu, Shuaicheng, Cao, Yang, Lim, Wei Yang Bryan
Format: Preprint
Published: 2024
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866909344262520832
author Zhang, Jingyuan
Duan, Yiyang
Niu, Shuaicheng
Cao, Yang
Lim, Wei Yang Bryan
author_facet Zhang, Jingyuan
Duan, Yiyang
Niu, Shuaicheng
Cao, Yang
Lim, Wei Yang Bryan
contents Federated Domain Adaptation (FDA) is a Federated Learning (FL) scenario where models are trained across multiple clients with unique data domains but a shared category space, without transmitting private data. The primary challenge in FDA is data heterogeneity, which causes significant divergences in gradient updates when using conventional averaging-based aggregation methods, reducing the efficacy of the global model. This further undermines both in-domain and out-of-domain performance (within the same federated system but outside the local client). To address this, we propose a novel framework called \textbf{M}ulti-domain \textbf{P}rototype-based \textbf{F}ederated Fine-\textbf{T}uning (MPFT). MPFT fine-tunes a pre-trained model using multi-domain prototypes, i.e., pretrained representations enriched with domain-specific information from category-specific local data. This enables supervised learning on the server to derive a globally optimized adapter that is subsequently distributed to local clients, without the intrusion of data privacy. Empirical results show that MPFT significantly improves both in-domain and out-of-domain accuracy over conventional methods, enhancing knowledge preservation and adaptation in FDA. Notably, MPFT achieves convergence within a single communication round, greatly reducing computation and communication costs. To ensure privacy, MPFT applies differential privacy to protect the prototypes. Additionally, we develop a prototype-based feature space hijacking attack to evaluate robustness, confirming that raw data samples remain unrecoverable even after extensive training epochs. The complete implementation of MPFL is available at \url{https://anonymous.4open.science/r/DomainFL/}.
format Preprint
id arxiv_https___arxiv_org_abs_2410_07738
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Enhancing Federated Domain Adaptation with Multi-Domain Prototype-Based Federated Fine-Tuning
Zhang, Jingyuan
Duan, Yiyang
Niu, Shuaicheng
Cao, Yang
Lim, Wei Yang Bryan
Machine Learning
Artificial Intelligence
Federated Domain Adaptation (FDA) is a Federated Learning (FL) scenario where models are trained across multiple clients with unique data domains but a shared category space, without transmitting private data. The primary challenge in FDA is data heterogeneity, which causes significant divergences in gradient updates when using conventional averaging-based aggregation methods, reducing the efficacy of the global model. This further undermines both in-domain and out-of-domain performance (within the same federated system but outside the local client). To address this, we propose a novel framework called \textbf{M}ulti-domain \textbf{P}rototype-based \textbf{F}ederated Fine-\textbf{T}uning (MPFT). MPFT fine-tunes a pre-trained model using multi-domain prototypes, i.e., pretrained representations enriched with domain-specific information from category-specific local data. This enables supervised learning on the server to derive a globally optimized adapter that is subsequently distributed to local clients, without the intrusion of data privacy. Empirical results show that MPFT significantly improves both in-domain and out-of-domain accuracy over conventional methods, enhancing knowledge preservation and adaptation in FDA. Notably, MPFT achieves convergence within a single communication round, greatly reducing computation and communication costs. To ensure privacy, MPFT applies differential privacy to protect the prototypes. Additionally, we develop a prototype-based feature space hijacking attack to evaluate robustness, confirming that raw data samples remain unrecoverable even after extensive training epochs. The complete implementation of MPFL is available at \url{https://anonymous.4open.science/r/DomainFL/}.
title Enhancing Federated Domain Adaptation with Multi-Domain Prototype-Based Federated Fine-Tuning
topic Machine Learning
Artificial Intelligence
url https://arxiv.org/abs/2410.07738