Abstract: Recent attention to privacy issues demands a communication-safe method for training human activity recognition (HAR) models on client activity data. Federated learning (FL) has become a compelling technique to facilitate model training between the server and clients while preserving data privacy. However, classical FL methods often assume independent and identically distributed (IID) data among clients. This assumption does not hold true in practical scenarios. Human activity in real-world scenarios varies, resulting in skewness where identical activities are executed uniquely across clients. This leads to local model objectives drifting away from the global model objective, thereby impeding overall convergence. To address this challenge, we propose FedCoad, a novel federated model leveraging contrastive learning with adaptive control variates to handle the skewness among HAR clients. Model contrastive learning minimizes the gap in representation between global and local models to help global model convergence. During local model updates, the adaptive control variates penalize the local model updates with respect to the model weight and the rate of change from the control variates update. Our experiments show that FedCoad outperforms state-of-the-art FL algorithms on HAR benchmark datasets.

用于人体活动识别的基于对比学习与自适应变量控制的联邦模型

[1]Bianchi V, Bassoli M, Lombardo G, et al., 2019. IoT wearable sensor and deep learning: an integrated approach for personalized human activity recognition in a smart home environment. IEEE Int Things J, 6(5):8553-8562.

[2]Diao EM, Ding J, Tarokh V, 2021. HeteroFL: computation and communication efficient federated learning for heterogeneous clients. Proc 9^th Int Conf on Learning Representations.

[3]Guo CR, Zhang YW, Chen YQ, et al., 2024. Modality consistency-guided contrastive learning for wearable-based human activity recognition. IEEE Int Things J, 11(12):21750-21762.

[4]Guo JL, Li ZT, Liu AF, et al., 2024. REC-Fed: a robust and efficient clustered federated system for dynamic edge networks. IEEE Trans Mob Comput, 23(12):15256-15273.

[5]Hsu TMH, Qi H, Brown M, 2019. Measuring the effects of non-identical data distribution for federated visual classification. https://arxiv.org/abs/1909.06335

[6]Karimireddy SP, Kale S, Mohri M, et al., 2020. SCAFFOLD: stochastic controlled averaging for federated learning. Proc 37^th Int Conf on Machine Learning, p.5132-5143.

[7]Kwapisz JR, Weiss GM, Moore SA, 2011. Activity recognition using cell phone accelerometers. ACM SIGKDD Explor Newsl, 12(2):74-82.

[8]Li CL, Niu D, Jiang B, et al., 2021. Meta-HAR: federated representation learning for human activity recognition. Proc Web Conf, p.912-922.

[9]Li QB, He BS, Song D, 2021. Model-contrastive federated learning. Proc IEEE/CVF Conf on Computer Vision and Pattern Recognition, p.10713-10722.

[10]Li T, Sahu AK, Zaheer M, et al., 2020. Federated optimization in heterogeneous networks. Proc 3^rd Conf on Machine Learning and Systems.

[11]Li X, Huang KX, Yang WH, et al., 2020. On the convergence of FedAvg on non-IID data. Proc 8^th Int Conf on Learning Representations.

[12]Malekzadeh M, Clegg RG, Cavallaro A, et al., 2019. Mobile sensor data anonymization. Proc Int Conf on Internet of Things Design and Implementation, p.49-58.

[13]McMahan HB, Moore E, Daniel R, et al., 2017. Communication-efficient learning of deep networks from decentralized data. Proc 20^th Int Conf on Artificial Intelligence and Statistics, p.1273-1282.

[14]Mrozek D, Koczur A, Małysiak-Mrozek B, 2020. Fall detection in older adults with mobile IoT devices and machine learning in the cloud and on the edge. Inform Sci, 537:132-147.

[15]Presotto R, Civitarese G, Bettini C, 2022. FedCLAR: federated clustering for personalized sensor-based human activity recognition. Proc IEEE Int Conf on Pervasive Computing and Communications, p.227-236.

[16]Sozinov K, Vlassov V, Girdzijauskas S, 2018. Human activity recognition using federated learning. Proc IEEE Int Conf on Parallel & Distributed Processing with Applications, Ubiquitous Computing & Communications, Big Data & Cloud Computing, Social Computing & Networking, Sustainable Computing & Communications, p.1103-1111.

[17]Stisen A, Blunck H, Bhattacharya S, et al., 2015. Smart devices are different: assessing and mitigating mobile sensing heterogeneities for activity recognition. Proc 13^th ACM Conf on Embedded Networked Sensor Systems, p.127-140.

[18]Tang CI, Perez-Pozuelo I, Spathis D, et al., 2021. SelfHAR: improving human activity recognition through self-training with unlabeled data. Proc ACM Interact Mob Wear Ubiq Technol, 5(1):36.

[19]Wang HY, Yurochkin M, Sun YK, et al., 2020. Federated learning with matched averaging. Proc 8^th Int Conf on Learning Representations.

[20]Wu JH, Sun J, Song J, et al., 2022. Health assessment method based on multi-sign information fusion of body area network. Inform Sci, 618:136-149.