Deep Neural Network Services for IoT Applications

Keynote speech by Tarek Abdelzaher, University of Illinois at Urbana Champaign
Nov. 10, 2019 at 9:15 - 10:15 am

Abstract: The talk discusses a suite of machine learning and inference services designed to endow simple embedded devices (the clients of the service) with the appearance of intelligent behavior. Much of the talk offers answers to the followingquestions: How much of the AI workflow can be done on the resource-limited and power-limited end-devices themselves? How to efficiently delegate the rest to a remote service? Short of understanding laws of nature, can neural networks be made more efficientat learning from sensor data that reflects processes in the physical world? Finally, with the emergence of safety-critical applications, such as autonomous cars, is the current AI workflow adequate to meet priority and criticality needs of such applications? Examples and evaluation results are also presented.

Bio: Tarek Abdelzaher received his Ph.D. in Computer Science from the University of Michigan in 1999. He is currently a Professor and Willett Faculty Scholar at the Department of Computer Science, the University of Illinois at Urbana Champaign.He has authored/coauthored more than 300 refereed publications in real-time computing, distributed systems, sensor networks, and control. He is an Editor-in-Chief of the Journal of Real-Time Systems, and has served as Associate Editor of the IEEE Transactionson Mobile Computing, IEEE Transactions on Parallel and Distributed Systems, IEEE Embedded Systems Letters, the ACM Transaction on Sensor Networks, and the Ad Hoc Networks Journal. He chaired (as Program or General Chair) several conferences in his area includingRTAS, RTSS, IPSN, Sensys, DCoSS, ICDCS, and ICAC. Abdelzaher’s research interests lie broadly in understanding and influencing performance and temporal properties of networked embedded, social and software systems in the face of increasing complexity, distribution,and degree of interaction with an external physical environment. Tarek Abdelzaher is a recipient of the IEEE Outstanding Technical Achievement and Leadership Award in Real-time Systems (2012), the Xerox Award for Faculty Research (2011), as well as severalbest paper awards. He is a member of IEEE and ACM.