Understanding Uncertainty of Edge Computing: New Principle and Design Approach

TL;DR

This paper explores the uncertainties in edge computing caused by dataset shift and introduces a new principle of AI model diversity, proposing MoDNet to leverage multiple models for improved uncertainty management.

Contribution

It introduces the principle of AI model diversity for edge computing and proposes MoDNet, a novel network architecture to exploit this diversity for better uncertainty handling.

Findings

Proposes the Model Diversity Network (MoDNet) for edge computing.

Highlights the importance of AI model diversity in managing dataset shift.

Provides design guidelines for learning-driven communication schemes.

Abstract

Due to the edge's position between the cloud and the users, and the recent surge of deep neural network (DNN) applications, edge computing brings about uncertainties that must be understood separately. Particularly, the edge users' locally specific requirements that change depending on time and location cause a phenomenon called dataset shift, defined as the difference between the training and test datasets' representations. It renders many of the state-of-the-art approaches for resolving uncertainty insufficient. Instead of finding ways around it, we exploit such phenomenon by utilizing a new principle: AI model diversity, which is achieved when the user is allowed to opportunistically choose from multiple AI models. To utilize AI model diversity, we propose Model Diversity Network (MoDNet), and provide design guidelines and future directions for efficient learning driven communication schemes.