用预训练模型实现轻量量化，降低物联网联邦学习能耗。
QuantFL: Sustainable Federated Learning for Edge IoT via Pre-Trained Model Quantisation
- 利用预训练模型初始化，支持高效低比特量化。
- 通信量减少40%以上，上行通信节省超80%。
- 适合电池受限的边缘IoT设备可持续训练。
联邦学习（FL）可在物联网（IoT）设备上实现隐私保护智能，但频繁上行传输导致显著碳足迹。尽管边缘设备越来越多部署预训练模型，其在微调阶段降低能耗的潜力尚未充分挖掘。本文提出QuantFL，一种可持续的联邦学习框架，通过预训练初始化实现激进且计算轻量的量化。我们证明预训练能自然集中更新统计信息，使我们可采用内存高效的桶量化，无需复杂误差反馈机制带来的高能耗。在MNIST和CIFAR-100上，QuantFL将总通信量减少40%（全精度下行时约40%总比特减少；上行或下行量化时≥80%），在严格带宽约束下仍达到或超过未压缩基线性能；在仅使用极少比特的情况下，测试准确率分别达89.00%（MNIST）和66.89%（CIFAR-100）。我们还分析了上行与下行通信成本，并对量化等级与初始化方式进行了消融实验。QuantFL为电池受限的物联网网络提供了实用的绿色可扩展训练方案。
原文摘要 · Abstract (English)
Federated Learning (FL) enables privacy-preserving intelligence on Internet of Things (IoT) devices but incurs a significant carbon footprint due to the high energy cost of frequent uplink transmission. While pre-trained models are increasingly available on edge devices, their potential to reduce the energy overhead of fine-tuning remains underexplored. In this work, we propose QuantFL, a sustainable FL framework that leverages pre-trained initialisation to enable aggressive, computationally lightweight quantisation. We demonstrate that pre-training naturally concentrates update statistics, allowing us to use memory-efficient bucket quantisation without the energy-intensive overhead of complex error-feedback mechanisms. On MNIST and CIFAR-100, QuantFL reduces total communication by 40\% ($\simeq40\%$ total-bit reduction with full-precision downlink; $\geq80\%$ on uplink or when downlink is quantised) while matching or exceeding uncompressed baselines under strict bandwidth budgets; BU attains 89.00\% (MNIST) and 66.89\% (CIFAR-100) test accuracy with orders of magnitude fewer bits. We also account for uplink and downlink costs and provide ablations on quantisation levels and initialisation. QuantFL delivers a practical, "green" recipe for scalable training on battery-constrained IoT networks.
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。