Distilling On-Device Intelligence at the Network Edge

• Published in: arXiv.org 2019-08
• Main Authors: Park, Jihong, Wang, Shiqiang, Elgabli, Anis, Oh, Seungeun, Jeong, Eunjeong, Cha, Han, Kim, Hyesung, Kim, Seong-Lyun, Bennis, Mehdi
• Format: Article
• Language: English
• Subjects: Autonomous cars; Cloud computing; Datasets; Distillation; Electronic devices; Machine learning; User requirements; Wireless networks
• ISSN: 2331-8422

Devices at the edge of wireless networks are the last mile data sources for machine learning (ML). As opposed to traditional ready-made public datasets, these user-generated private datasets reflect the freshest local environments in real time. They are thus indispensable for enabling mission-critical intelligent systems, ranging from fog radio access networks (RANs) to driverless cars and e-Health wearables. This article focuses on how to distill high-quality on-device ML models using fog computing, from such user-generated private data dispersed across wirelessly connected devices. To this end, we introduce communication-efficient and privacy-preserving distributed ML frameworks, termed fog ML (FML), wherein on-device ML models are trained by exchanging model parameters, model outputs, and surrogate data. We then present advanced FML frameworks addressing wireless RAN characteristics, limited on-device resources, and imbalanced data distributions. Our study suggests that the full potential of FML can be reached by co-designing communication and distributed ML operations while accounting for heterogeneous hardware specifications, data characteristics, and user requirements.