针对联邦语言模型微调，只更新关键层提升收敛与性能
FedTLU: Federated Learning with Targeted Layer Updates
- 按重要性评分选择性更新语言模型层，其余层冻结
- 在非独立同分布数据下，收敛速度与准确率显著提升
- 适合数据异质性强的联邦学习场景，提升模型鲁棒性
联邦学习（FL）通过允许多个客户端参与训练而无需共享数据，缓解了语言模型训练中的隐私问题。然而，客户端间数据分布不一致（non-IID）常限制FL性能，尤其在模型微调阶段，数据差异带来的噪声会损害模型在稳定点附近的收敛。本文提出一种面向微调的有目标层更新策略：不再随机更新语言模型各层，而是通过评分机制识别并更新最关键的层，其余层保持冻结，以避免过度噪声或恶意更新。大量实验表明，该方法在非IID设置下显著提升收敛性与模型性能，为联邦语言模型微调提供更高效方案。
原文摘要 · Abstract (English)
Federated learning (FL) addresses privacy concerns in training language models by enabling multiple clients to contribute to the training, without sending their data to others. However, non-IID (identically and independently distributed) data across clients often limits FL's performance. This issue is especially challenging during model fine-tuning, as noise due to variations in clients' data distributions can harm model convergence near stationary points. This paper proposes a targeted layer update strategy for fine-tuning in FL. Instead of randomly updating layers of the language model, as often done in practice, we use a scoring mechanism to identify and update the most critical layers, avoiding excessively noisy or even poisoned updates by freezing the parameters in other layers. We show in extensive experiments that our method improves convergence and performance in non-IID settings, offering a more efficient approach to fine-tuning federated language models.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。