Edge computing has been emerged as attractive attention and possesses several challenges relating to shared communication and computation resources across multiple edge devices. Edge nodes are resource-restrained compared to the cloud. The energy efficiency of deep learning models is important as IoT, or edge devices have limited computation and energy resources.
Conventional Deep learning models employ optimize and schedule resources in edge computing networks. Although, when deploying heavier Deep Neural Network models for computational tasks in an edge computing platform, it is the lack to support the execution of computationally intensive tasks due to limited computing resources in the mobile web devices. Thus, with the help of promising edge computing, a lightweight deep neural network for mobile web devices helps to improve energy efficiency effectively.