Mobile sensing through deep learning

Abstract

Today, mobile devices are equipped with powerful processors along with various on-device sensors. Over the past few years, deep learning has become the dominant approach in the field of machine learning due to its impressive performance. We envision that in the near future, powered by deep learning, mobile devices will become more intelligent and revolutionize a wide range of applications. In this paper, we discuss the challenges of enabling deep learning on mobile platforms. Our work is to propose a deep learning framework that achieves state-of-the-art performance with low overhead on resource-limited mobile platforms. Our preliminary results show that deep learning can efficiently solve object recognition problem under noisy real world environment.