EdgeRL: Reinforcement Learning-driven Deep Learning Model Inference Optimization at Edge

TL;DR

This paper proposes EdgeRL framework that seeks to strike such balance by using an Advantage Actor-Critic (A2C) Reinforcement Learning (RL) approach that can choose optimal run-time DNN inference parameters and aligns the performance metrics based on the application requirements.

Abstract

Balancing mutually diverging performance metrics, such as, processing latency, outcome accuracy, and end device energy consumption is a challenging undertaking for deep learning model inference in ad-hoc edge environments. In this paper, we propose EdgeRL framework that seeks to strike such balance by using an Advantage Actor-Critic (A2C) Reinforcement Learning (RL) approach that can choose optimal run-time DNN inference parameters and aligns the performance metrics based on the application requirements. Using real world deep learning model and a hardware testbed, we evaluate the benefits of EdgeRL framework in terms of end device energy savings, inference accuracy improvement, and end-to-end inference latency reduction.

Figures (4)

• Figure 1: Ad-hoc edge deployment and potential implementation of the EdgeRL framework
• Figure 2: System performance over varying accuracy weight
• Figure 3: System performance over varying latency weight
• Figure 4: System performance over varying energy weight