New Directions in Distributed Deep Learning: Bringing the Network at Forefront of IoT Design

TL;DR

This paper discusses the challenges of deploying deep learning on IoT devices and proposes three research directions—federated learning, data-independent algorithms, and communication-aware inference—to enable network-centric edge intelligence.

Contribution

It introduces a unified view of three emerging research directions addressing hardware, privacy, and network challenges in distributed deep learning for IoT.

Findings

Identifies key challenges in IoT deep learning deployment.

Proposes three research directions for network-centric edge intelligence.

Highlights the importance of communication-aware distributed inference.

Abstract

In this paper, we first highlight three major challenges to large-scale adoption of deep learning at the edge: (i) Hardware-constrained IoT devices, (ii) Data security and privacy in the IoT era, and (iii) Lack of network-aware deep learning algorithms for distributed inference across multiple IoT devices. We then provide a unified view targeting three research directions that naturally emerge from the above challenges: (1) Federated learning for training deep networks, (2) Data-independent deployment of learning algorithms, and (3) Communication-aware distributed inference. We believe that the above research directions need a network-centric approach to enable the edge intelligence and, therefore, fully exploit the true potential of IoT.