… His research interests include the Internet of Things, ubiquitous computing, mobile computing,
embedded AI, wireless sensor networks, and big data analytics. He is currently serving as …

… MDLdroid, a novel decentralized Mobile Deep Learning framework to enable resource-aware
on-device collaborative learning for personal mobile sensing … MDLdroid targets to fully …

… on-device learning. In this paper, we present a novel on-device deep learning framework
named MDL… Alternatively, existing MDL frameworks, such as FL [30] and MDLdroid [63], can be …

… , we propose a ChainSGDreduce approach which … MDLdroid on off-the-shelf Android
smartphones with a number of state-of-the-art DL models using 6 public personal mobile sensing …

… TagAlong: Efficient Integration of Battery-Free Sensor Tags in Standard Wireless Networks
… Deep Learning in Signal Translation for Cross Configurations in Device-Free WiFi Sensing …

… applies the differential privacy in a mobile cloud deep learning framework [6], and we use …
MDLdroid: A ChainSGD-reduce approach to mobile deep learning for personal mobile sensing…

… mobile users, UAVs/BSs, and a heterogeneous computing platform. In the FL architectures,
each mobile user holds a deep neural network … A testbed is implemented in a wireless sensor …

… Deep learning-based visual sensing has achieved attractive … ], a ChainSGD-reduce algorithm
and reinforcement learning … efficient deep models on embedded and mobile devices [36…

… manufacturers to train a machine learning model based on … Specifically, we use a mobile
phone to collect data from home … for wireless communication, Internet of Things, and sensor …

… utilize sensors in mobile devices and machine learning to … domain) a critical issue in mobile
sensing. Despite attempts in domain … Our evaluation with four real-world sensing datasets …