An Offloading Algorithm for Maximizing Inference Accuracy on Edge Device in an Edge Intelligence System

Abstract

With the emergence of edge computing, the problem of offloading jobs between an Edge Device (ED) and an Edge Server (ES) received significant attention in the past. Motivated by the fact that an increasing number of applications are using Machine Learning (ML) inference from the data samples collected at the EDs, we study the problem of offloading inference jobs by considering the following novel aspects: in contrast to a typical computational job 1) both inference accuracy and processing time of an inference job increase with the size of the ML model and 2)recently proposed Deep Neural Networks (DNNs) for resource-constrained EDs provide the choice of scaling down the model size by trading off the inference accuracy. Therefore, we consider that multiple small-size ML models are available at the ED and a powerful large-size ML model is available at the ES, and study a general assignment problem with the objective of maximizing the total inference accuracy for the data samples at the ED subject to a time constraint T on the makespan. Noting that the problem is NP-hard, we propose an approximation algorithm: Accuracy Maximization using LP-Relaxation and Rounding (AMR2), and prove that it results in a makespan at most 2T, and achieves a total accuracy that is lower by a small constant from the optimal total accuracy. As proof of concept, we implemented AMR2 on a Raspberry Pi, equipped with MobileNets, that is connected via LAN to a server equipped with ResNet, and studied the total accuracy and makespan performance of AMR2 for image classification.