Federated Fine-Tuning for Pre-Trained Foundation Models Over Wireless Networks

TL;DR

This paper introduces a split federated LoRA framework for efficient, privacy-preserving fine-tuning of pre-trained foundation models over wireless networks, addressing resource constraints and optimizing device scheduling.

Contribution

It proposes a novel split federated LoRA framework with theoretical convergence analysis and an online algorithm for device scheduling and bandwidth allocation in wireless settings.

Findings

The split federated LoRA framework achieves comparable performance to full fine-tuning.

The online algorithm improves learning efficiency under resource constraints.

Simulation results validate the effectiveness of the proposed approach.

Abstract

Pre-trained foundation models (FMs), with extensive number of neurons, are key to advancing next-generation intelligence services, where personalizing these models requires massive amount of task-specific data and computational resources. The prevalent solution involves centralized processing at the edge server, which, however, raises privacy concerns due to the transmission of raw data. Instead, federated fine-tuning (FedFT) is an emerging privacy-preserving fine-tuning (FT) paradigm for personalized pre-trained foundation models. In particular, by integrating low-rank adaptation (LoRA) with federated learning (FL), federated LoRA enables the collaborative FT of a global model with edge devices, achieving comparable learning performance to full FT while training fewer parameters over distributed data and preserving raw data privacy. However, the limited radio resources and computation capabilities of edge devices pose significant challenges for deploying federated LoRA over wireless networks. To this paper, we propose a split federated LoRA framework, which deploys the computationally-intensive encoder of a pre-trained model at the edge server, while keeping the embedding and task modules at the edge devices. Building on this split framework, the paper provides a rigorous analysis of the upper bound of the convergence gap for the wireless federated LoRA system. This analysis motivates the formulation of a long-term upper bound minimization problem, where we decompose the formulated long-term mixed-integer programming (MIP) problem into sequential sub-problems using the Lyapunov technique. We then develop an online algorithm for effective device scheduling and bandwidth allocation. Simulation results demonstrate the effectiveness of the proposed online algorithm in enhancing learning performance.