Energy-Efficient Split Learning for Fine-Tuning Large Language Models in Edge Networks

TL;DR

This paper introduces an energy-efficient split learning framework for fine-tuning large language models at the network edge, optimizing delay and energy use across heterogeneous devices.

Contribution

It proposes the CARD algorithm to minimize training delay and energy consumption considering device heterogeneity and channel dynamics.

Findings

Reduces training delay by 70.8%

Lowers server energy consumption by 53.1%

Effective for geo-distributed edge networks

Abstract

In this letter, we propose an energy-efficient split learning (SL) framework for fine-tuning large language models (LLMs) using geo-distributed personal data at the network edge, where LLMs are split and alternately across massive mobile devices and an edge server. Considering the device heterogeneity and channel dynamics in edge networks, a \underline{C}ut l\underline{A}yer and computing \underline{R}esource \underline{D}ecision (CARD) algorithm is developed to minimize training delay and energy consumption. Simulation results demonstrate that the proposed approach reduces the average training delay and server's energy consumption by 70.8% and 53.1%, compared to the benchmarks, respectively.