Optimizing Deep Learning Application for Edge Computing

Abstract

eep learning (DL) models such as convolutional neural networks (CNN) are being deployed to solve various computer vision and natural language processing tasks at the edge. It is a challenge to find the right DL architecture that simultaneously meets the accuracy, power and performance budgets of such resource-constrained devices. Hardware-aware Neural Architecture Search (HW-NAS) has recently gained steam by automating the design of efficient DL models for a variety of target hardware platform.However, such algorithms require excessive computational resources. Thousands of GPU days are required to evaluate and explore modern DL architecture search space. In this talk I will present state-of-the-art approaches that are based on two components: a) Surrogate models to predict quickly architecture accuracy and hardware performances to speed up HW-NAS, b) Efficient search algorithm that explores only promising hardware and software regions of the search space.