pFedLoRA: Model-Heterogeneous Personalized Federated Learning with LoRA Tuning
This work addresses efficiency issues in federated learning for LLMs, offering a solution for scenarios with model heterogeneity, though it is incremental in adapting existing LoRA methods to this context.
The paper tackles the challenge of high computational and communication costs in model-heterogeneous personalized federated learning (MHPFL) when using large language models (LLMs), proposing pFedLoRA, which uses homogeneous small adapters for efficient training. It demonstrates improvements, including a 1.35% increase in test accuracy, 11.81 times computation reduction, and 7.41 times communication cost saving compared to baselines.
Federated learning (FL) is an emerging machine learning paradigm in which a central server coordinates multiple participants (clients) collaboratively to train on decentralized data. In practice, FL often faces statistical, system, and model heterogeneities, which inspires the field of Model-Heterogeneous Personalized Federated Learning (MHPFL). With the increased interest in adopting large language models (LLMs) in FL, the existing MHPFL methods cannot achieve acceptable computational and communication costs, while maintaining satisfactory model performance. To bridge this gap, we propose a novel and efficient model-heterogeneous personalized Federated learning framework based on LoRA tuning (pFedLoRA). Inspired by the popular LoRA method for fine-tuning pre-trained LLMs with a low-rank model (a.k.a., an adapter), we design a homogeneous small adapter to facilitate federated client's heterogeneous local model training with our proposed iterative training for global-local knowledge exchange. The homogeneous small local adapters are aggregated on the FL server to generate a global adapter. We theoretically prove the convergence of pFedLoRA. Extensive experiments on two benchmark datasets demonstrate that pFedLoRA outperforms six state-of-the-art baselines, beating the best method by 1.35% in test accuracy, 11.81 times computation overhead reduction and 7.41 times communication cost saving.