2 March 2021 in IEEE Access
The ever-increasing number of Internet of Things (IoT) devices are continuously generating huge masses of data, but the current cloud-centric approach for IoT big data analysis has raised public concerns on both data privacy and network cost. Federated learning (FL) recently emerges as a promising technique to accommodate these concerns, by means of learning a global model by aggregating local updates from multiple devices without sharing the privacy-sensitive data. However, IoT devices usually have constrained computation resources and poor network connections, making it infeasible or very slow to train deep neural networks (DNNs) by following the FL pattern. To address this problem, we propose a new efficient FL framework called FL-PQSU in this paper. It is composed of 3-stage pipeline: structured pruning, weight quantization and selective updating, that work together to reduce the costs of computation, storage, and communication to accelerate the FL training process. We study FL-PQSU using popular DNN models (AlexNet, VGG16) and publicly available datasets (MNIST, CIFAR10), and demonstrate that it can well control the training overhead while still guaranteeing the learning performance.
23 February 2023 in ACM Transactions on Internet Technology
The advent of Deep Neural Networks (DNNs) has empowered numerous computer-vision applications. Due to the high computational intensity of DNN models, as well as the resource constrained nature of Industrial Internet-of-Things (IIoT) devices, it is generally very challenging to deploy and execute DNNs efficiently in the industrial scenarios. Substantial research has focused on model compression or edge-cloud offloading, which trades off accuracy for efficiency or depends on high-quality infrastructure support, respectively. In this article, we present EdgeDI, a framework for executing DNN inference in a partitioned, distributed manner on a cluster of IIoT devices. To improve the inference performance, EdgeDI exploits two key optimization knobs, including: (1) Model compression based on deep architecture design, which transforms the target DNN model into a compact one that reduces the resource requirements for IIoT devices without sacrificing accuracy; (2) Distributed inference based on adaptive workload partitioning, which achieves high parallelism by adaptively balancing the workload distribution among IIoT devices under heterogeneous resource conditions. We have implemented EdgeDI based on PyTorch, and evaluated its performance with the NEU-CLS defect classification task and two typical DNN models (i.e., VGG and ResNet) on a cluster of heterogeneous Raspberry Pi devices. The results indicate that the proposed two optimization approaches significantly outperform the existing solutions in their specific domains. When they are well combined, EdgeDI can provide scalable DNN inference speedups that are very close to or even much higher than the theoretical speedup bounds, while still maintaining the desired accuracy.