Abstract:Recent breakthrough technological progressions of powerful mobile computing resources such as low-cost mobile GPUs along with cutting-edge, open-source software architectures have enabled high-performance deep learning on mobile platforms. These advancements have revolutionized the capabilities of today's mobile applications in different dimensions to perform data-driven intelligence locally, particularly for smart health applications. Unlike traditional machine learning (ML) architectures, modern on-device deep learning frameworks are proficient in utilizing computing resources in mobile platforms seamlessly, in terms of producing highly accurate results in less inference time. However, on the flip side, energy resources in a mobile device are typically limited. Hence, whenever a complex Deep Neural Network (DNN) architecture is fed into the on-device deep learning framework, while it achieves high prediction accuracy (and performance), it also urges huge energy demands during the runtime. Therefore, managing these resources efficiently within the spectrum of performance and energy efficiency is the newest challenge for any mobile application featuring data-driven intelligence beyond experimental evaluations. In this paper, first, we provide a timely review of recent advancements in on-device deep learning while empirically evaluating the performance metrics of current state-of-the-art ML architectures and conventional ML approaches with the emphasis given on energy characteristics by deploying them on a smart health application. With that, we are introducing a new framework through an energy-aware, adaptive model comprehension and realization (EAMCR) approach that can be utilized to make more robust and efficient inference decisions based on the available computing/energy resources in the mobile device during the runtime.

DataLearner: A Data Mining and Knowledge Discovery Tool for Android Smartphones and Tablets

Close the Gap Between Deep Learning and Mobile Intelligence by Incorporating Training in the Loop

A Novel Personal Health Assessment Based on Mobile Telecom Data

Explore Training of Deep Convolutional Neural Networks on Battery-powered Mobile Devices: Design and Application

A Mobile English Learning Platform Based on Data Mining and Personalized Recommendations

Mobile Big Data Analytics Using Deep Learning and Apache Spark

"Know" Your Mobile Customers: A Design Approach to an Android-Based Mobile Analytics Tool.

Towards Tool-Support for Interactive-Machine Learning Applications in the Android Ecosystem

Droiddetector: Android Malware Characterization and Detection Using Deep Learning

Towards Implementing Energy-aware Data-driven Intelligence for Smart Health Applications on Mobile Platforms

A Non-Intrusive Machine Learning Solution for Malware Detection and Data Theft Classification in Smartphones

MobileMiner

Distributed Learning on Mobile Devices: A New Approach to Data Mining in the Internet of Things

Android Spyware Detection Using Machine Learning: A Novel Dataset

Exploring Mobile Data on Smartphones from Collection to Analysis

A First Look at Deep Learning Apps on Smartphones

Real-World Relevant Learning with Android Smartphones

Deep Learning Towards Mobile Applications

Smart Diary: A Smartphone-Based Framework for Sensing, Inferring, and Logging Users’ Daily Life

Mobileminer: A Real World Case Study Of Data Mining In Mobile Communication

DataLab: A Platform for Data Analysis and Intervention