Friday, October 11, 2019
Edge Intelligence in the Making: A Collaborative Learning Approach
Junshan Zhang, Arizona State University
11:00am, DBH 6011
[Hosts: Athina, Hamid]
Many IoT applications demand intelligent decisions in a real-time manner. The necessity of real-time edge intelligence dictates that decision making takes place right here right now at the network edge. Since an edge node often has a limited amount of data and is constrained with computational resources, we advocate collaborative learning to achieve edge intelligence. To this end, we first develop an edge learning framework based on collaboration between the edge node and the cloud, where the edge node learns its model based on local data in real-time, while leveraging the cloud intelligence. Specifically, the knowledge transfer from the cloud to the edge node is in the form of a reference model and its associated uncertainty set, and the edge node further constructs an uncertainty set centered around its local empirical distribution. The edge learning problem is then cast as a distributionally robust optimization (DRO) problem subject to the two distribution uncertainty sets, and the solution is characterized using Optimal Transport Theory. Next, we propose a federated meta-learning framework where multiple edge nodes collaboratively train the meta-learning model such that the obtained model can be quickly adapted by the target edge node, using its local data, to achieve real-time edge intelligence. We analyze the convergence of the federated meta-learning algorithm and examine the learning performance of the meta-trained model at the target edge. To the best of our knowledge, this is the first work to establish the convergence of meta-learning.
Junshan Zhang received his Ph.D. degree from the School of ECE at Purdue University in 2000. He joined the School of ECEE at Arizona State University in August 2000, where he has been Fulton Chair Professor since 2015. His research interests fall in the general field of information networks and data science, including communication networks, machine learning for Internet of Things (IoT), Fog/edge Computing, optimization/control of cyber physical systems, smart grid. He is a Fellow of the IEEE, and a recipient of the ONR Young Investigator Award in 2005 and the NSF CAREER award in 2003. His papers have won a few awards, including the Best Student paper at WiOPT 2018, the Kenneth C. Sevcik Outstanding Student Paper Award of ACM SIGMETRICS 2016, the Best Paper Runner-up Award of IEEE INFOCOM 2009 and IEEE INFOCOM 2014, and the Best Paper Award at IEEE ICC 2008 and ICC 2017. Building on his research findings, he co-founded Smartiply Inc in 2015, a Fog Computing startup company delivering boosted network connectivity and embedded artificial intelligence. Prof. Zhang was TPC co-chair for a few major conferences in communication networks, including IEEE INFOCOM 2012 and ACM MOBIHOC 2015. He was general chair for ACM/IEEE SEC 2017 and WiOPT 2016.