为可穿戴健身追踪设计联邦学习基线，兼顾隐私与个性化。
FedFitTech: A Baseline in Federated Learning for Fitness Tracking
- 基于Flower框架构建联邦学习基线，支持跨设备协作训练。
- 通过客户端早停策略降低13%冗余通信，性能仅下降1%。
- 适合关注隐私保护、个性化建模的健康科技研究者。
传感器与轻量级机器学习模型的快速发展推动了可穿戴健身追踪设备的普及。这类设备搭载惯性传感器，可连续采集运动数据，应用于体育优化与预防性医疗。传统集中式学习面临数据隐私、监管限制和通信效率问题。相比之下，联邦学习（FL）通过交换模型更新而非原始数据实现去中心化训练。将FL应用于健身技术（FitTech）存在数据不平衡、标签缺失、用户活动异质性及个性化与泛化权衡等挑战。为此，本文提出公开可用的FedFitTech基线，基于Flower框架，已被学术界与工业界广泛采用。案例研究展示了基于该基线的系统，引入客户端早停策略，实现了常见动作识别与个体差异保留的平衡。实验表明，该方法减少13%的整体冗余通信，同时识别性能仅下降1%，显著提升隐私友好型健身追踪的可扩展性与效率。项目已开源：https://github.com/shreyaskorde16/FedFitTech。
原文摘要 · Abstract (English)
The rapid evolution of sensors and resource-efficient machine learning models has spurred the widespread adoption of wearable fitness tracking devices. Equipped with inertial sensors, such devices can continuously capture physical movements for fitness technology (FitTech), enabling applications from sports optimization to preventive healthcare. Traditional Centralized Learning approaches to detect fitness activities struggle with data privacy concerns, regulatory restrictions, and communication inefficiencies. In contrast, Federated Learning (FL) enables a decentralized model training by communicating model updates rather than potentially private wearable sensor data. Applying FL to FitTech presents unique challenges, such as data imbalance, lack of labeled data, heterogeneous user activities, and trade-offs between personalization and generalization. To simplify research on FitTech in FL, we present the FedFitTech baseline, under the Flower framework, which is publicly available and widely used by both industry and academic researchers. Additionally, to illustrate its usage, this paper presents a case study that implements a system based on the FedFitTech baseline, incorporating a client-side early stopping strategy and comparing the results. For instance, this system allows wearable devices to optimize the trade-off between capturing common fitness activities and preserving individuals' nuances, thereby enhancing both the scalability and efficiency of privacy-aware fitness tracking applications. The results show that this reduces the overall redundant communications by 13%, while maintaining the overall recognition performance at a negligible recognition cost by 1%. Thus, the FedFitTech baseline creates a foundation for a wide range of new research and development opportunities in FitTech, and it is available as open source at: https://github.com/shreyaskorde16/FedFitTech
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。