Fog Computing

3Deep Learning in the Era of Edge Computing: Challenges and Opportunities

Mi Zhang¹, Faen Zhang², Nicholas D. Lane³, Yuanchao Shu⁴, Xiao Zeng¹, Biyi Fang¹, Shen Yan¹, and Hui Xu²

¹Department of Electrical and Computer Engineering, Michigan State University, East Lansing, MI, USA, 48824

²AInnovation, Beijing, China, 100080

³Department of Computer Science, Oxford University, Oxford, United Kingdom, OX1 3PR

⁴Microsoft Research, Redmond, WA, USA, 98052

3.1 Introduction

Of all the technology trends that are taking place right now, perhaps the biggest one is edge computing [1, 2]. It is the one that is going to bring the most disruption and the most opportunity over the next decade. Broadly speaking, edge computing is a new computing paradigm that aims to leverage devices that are deployed at the Internet's edge to collect information from individuals and the physical world as well as to process the collected information in a distributed manner [3]. These devices, referred to as edge devices, are physical devices equipped with sensing, computing, and communication capabilities. Today, we are already surrounded by a variety of such edge devices: our mobile phones and wearables are edge devices; home intelligence devices such as Google Nest and Amazon Echo are edge devices; autonomous systems such as drones, self-driving vehicles, and robots that vacuum the carpet are also edge devices. These edge devices continuously collect a variety of data, including images, videos, audios, texts, ...