Proceedings of the 1st International Workshop on Deep Learning for Mobile Systems and Applications

Abstract

It is with true pleasure that we welcome you to the First International Workshop on Embedded and Mobile Deep Learning. This is a first-of-its-kind workshop for the mobile computing and systems community, and aims to offer a forum for both academics and practitioners alike to explore, discuss, and debate early findings and new vectors of research within this exciting area of machine learning as it applies to embedded and mobile systems. This workshop is largely motivated by the breakthroughs seen in the field of deep learning that are transforming how sensor data (e.g., images, audio, and even accelerometers and GPS) is modeled during the extraction of high-level information needed by the latest in sensor-driven systems like smartphone apps and wearable devices. Today, the state-of-the-art in computational models that, for example, recognize a face, track user emotions, or monitor physical activities are increasingly based on deep learning principles and algorithms. Unfortunately, deep models typically exert severe demands on local device resources and this conventionally limits their adoption within mobile and embedded platforms. As a result, in far too many cases existing systems process sensor data with machine learning methods that have been superseded by deep learning years ago. As the robustness and quality of sensory perception and reasoning is critical to mobile computing, it is important for this community to begin the careful study of two core technical questions. First, how should deep learning learning principles and algorithms be applied to sensor inference problems that are central to this class of computing? This includes a combination of applications of learning some of which are familiar to other domains (such as the processing image and audio), in addition to those more uniquely tied to wearable and mobile systems (e.g., activity recognition). Second, what is required for current -- and future -- deep learning advances to be either simplified or efficiently integrated into a variety of mobile resource-constrained systems? The scope of this workshop at MobiSys 2017 on embedded and mobile deep learning spans these two broad themes.