02578nas a2200313   4500000000100000000000100001008004100002260001200043653002300055653003900078653002500117653002600142653002300168653001900191653001600210653003700226653003200263100002300295700002100318700001900339700001800358700001800376245011800394856005800512300000900570490001300579520165800592022001402250       9998                            d  c12/202410aCommunication Cost10aConvolutional Neural Network (CNN)10aDeep Neural Networks10aDistributive Learning10aFederated Learning10aNeural Network10aPerformance10aResidual Neural Network (ResNet)10aVisual Geometry Group (VGG)1 aBasmah K. Alotaibi1 aFakhri Alam Khan1 aYousef Qawqzeh1 aGwanggil Jeon1 aDavid Camacho00aPerformance and Communication Cost of Deep Neural Networks in Federated Learning Environments: An Empirical Study  uhttps://www.ijimai.org/journal/bibcite/reference/3520  a1-120 vIn press3 aFederated learning, a distributive cooperative learning approach, allows clients to train the model locally using their data and share the trained model with a central server. When developing a federated learning environment, a deep/machine learning model needs to be chosen. The choice of the learning model can impact the model performance and the communication cost since federated learning requires the model exchange between clients and a central server in several rounds. In this work, we provide an empirical study to investigate the impact of using three different neural networks (CNN, VGG, and ResNet) models in image classification tasks using two different datasets (Cifar-10 and Cifar-100) in a federated learning environment. We investigate the impact of using these models on the global model performance and communication cost under different data distribution that are IID data and non-IID data distribution. The obtained results indicate that using CNN and ResNet models provide a faster convergence than VGG model. Additionally, these models require less communication costs. In contrast, the VGG model necessitates the sharing of numerous bits over several rounds to achieve higher accuracy under the IID data settings. However, its accuracy level is lower under non-IID data distributions than the other models. Furthermore, using a light model like CNN provides comparable results to the deeper neural network models with less communication cost, even though it may require more communication rounds to achieve the target accuracy in both datasets. CNN model requires fewer bits to be shared during communication than other models.  a1989-1660