As the development of the Internet of Things (IoT) continues, Federated Learning (FL) is gaining popularity as a distributed machine learning framework that does not compromise the data privacy of each participant. However, the data held by enterprises and factories in the IoT often have different distribution properties (Non-IID), leading to poor results in their federated learning. This problem causes clients to forget about global knowledge during their local training phase and then tends to slow convergence and degrades accuracy. In this work, we propose a method named FedRAD, which is based on relational knowledge distillation that further enhances the mining of high-quality global knowledge by local models from a higher-dimensional perspective during their local training phase to better retain global knowledge and avoid forgetting. At the same time, we devise an entropy-wise adaptive weights module (EWAW) to better regulate the proportion of loss in single-sample knowledge distillation versus relational knowledge distillation so that students can weigh losses based on predicted entropy and learn global knowledge more effectively. A series of experiments on CIFAR10 and CIFAR100 show that FedRAD has better performance in terms of convergence speed and classification accuracy compared to other advanced FL methods.

More from our Archive

• DOI: 10.3390/en16155682 2023

  Summary of Health-State Estimation of Lithium-Ion Batteries Based on Electrochemical Impedance Spectroscopy

  Xinwei Sun, Yang Zhang, Yongcheng Zhang, Licheng Wang, Kai Wang

• DOI: 10.3390/en16155809 2023

  Review of Family-Level Short-Term Load Forecasting and Its Application in Household Energy Management System

  Ping Ma, Shuhui Cui, Mingshuai Chen, Shengzhe Zhou, Kai Wang

• DOI: 10.3390/s23156949 2023

  Visual Intelligence in Precision Agriculture: Exploring Plant Disease Detection via Efficient Vision Transformers

  Sana Parez, Naqqash Dilshad, Norah Saleh Alghamdi, Turki M. Alanazi, Jong Weon Lee

• DOI: 10.3390/s23156762 2023

  HealthLock: Blockchain-Based Privacy Preservation Using Homomorphic Encryption in Internet of Things Healthcare Applications

  Aitizaz Ali, Bander Ali Saleh Al-rimy, Faisal S. Alsubaei, Abdulwahab Ali Almazroi, Abdulaleem Ali Almazroi

• DOI: 10.3390/machines11080774 2023

  Sustainable Crop Protection via Robotics and Artificial Intelligence Solutions

  Vasiliki Balaska, Zoe Adamidou, Zisis Vryzas, Antonios Gasteratos

• DOI: 10.3390/mi14071480 2023

  A Review of the Residual Stress Generation in Metal Additive Manufacturing: Analysis of Cause, Measurement, Effects, and Prevention

  Nabin Bastola, Muhammad P. Jahan, Nithin Rangasamy, Chandra Sekhar Rakurty

• DOI: 10.3390/electronics12143155 2023

  In-Memory Computing Integrated Structure Circuit Based on Nonvolatile Flash Memory Unit

  Peilong Xu, Dan Lan, Fengyun Wang, Incheol Shin

• DOI: 10.3390/s23146507 2023

  CNN-CNN: Dual Convolutional Neural Network Approach for Feature Selection and Attack Detection on Internet of Things Networks

  Basim Alabsi, Mohammed Anbar, Shaza Rihan

• DOI: 10.1039/d3mh00882g 2023

  Defect-stabilized and oxygen-coordinated iron single-atom sites facilitate hydrogen peroxide electrosynthesis

  Taotao Gao, Lu Qiu, Minghao Xie, Zhaoyu Jin, Panpan Li, Guihua Yu

• DOI: 10.3390/electronics12153300 2023

  Anomaly Detection in 6G Networks Using Machine Learning Methods

  Mamoon M. Saeed, Rashid A. Saeed, Maha Abdelhaq, Raed Alsaqour, Mohammad Kamrul Hasan, Rania A. Mokhtar