2020/08/25 by Kartikeya Bhardwaj, Wei Chen, Bhardwaj, Kartikeya +3
Computer Science · #Age of Information Optimization #Computer Vision and Pattern Recognition (cs.CV) #FOS: Computer and information sciences #FOS: Electrical engineering #IoT and Edge/Fog Computing #Machine Learning (cs.LG) #Machine Learning (stat.ML) #Privacy-Preserving Technologies in Data #Signal Processing (eess.SP) #electronic engineering #information engineering
paper · pdf · doi:10.48550/arxiv.2008.10805
openalex publication_date 2020/08/25 · openalex created_date 2022/07/26 · openalex updated_date 2026/07/28
In this paper, we first highlight three major challenges to large-scale\nadoption of deep learning at the edge: (i) Hardware-constrained IoT devices,\n(ii) Data security and privacy in the IoT era, and (iii) Lack of network-aware\ndeep learning algorithms for distributed inference across multiple IoT devices.\nWe then provide a unified view targeting three research directions that\nnaturally emerge from the above challenges: (1) Federated learning for training\ndeep networks, (2) Data-independent deployment of learning algorithms, and (3)\nCommunication-aware distributed inference. We believe that the above research\ndirections need a network-centric approach to enable the edge intelligence and,\ntherefore, fully exploit the true potential of IoT.\n