报告题目：Neural-enhanced Edge: Towards Operationalized AI in Cooperative Cloud-Edge Ecosystems
报告人简介：Song Guo is a Full Professor at Department of Computing, The Hong Kong Polytechnic University. He also holds a Changjiang Chair Professorship awarded by the Ministry of Education of China. His research interests are mainly in federated learning, edge AI, mobile computing, and distributed systems. He has published many papers in top venues and been consistently listed in the Clarivate Highly Cited Researchers. He is the recipient of over a dozen Best Paper Awards from IEEE/ACM conferences, journals, and technical committees. Prof. Guo is the Editor-in-Chief of IEEE Open Journal of the Computer Society (CS). He was a Distinguished Lecturer of IEEE Communications Society (ComSoc) and a member of IEEE ComSoc Board of Governors. He has served on IEEE Fellow Evaluation Committees for both CS and ComSoc, and been named on editorial board of a number of prestigious international journals like IEEE TC, TPDS, TCC, etc. Prof. Guo is a Fellow of the Canadian Academy of Engineering, Member of Academia Europaea, and Fellow of the IEEE.
报告摘要：As to paving the last mile of enabling intelligent applications on ubiquitous edge devices, it is necessary to deploy the entire lifecycle (ecosystem) of data generation, representation extraction and model consumption near the user side, with the support of diverse machine learning techniques. This emerging scenario promotes the rise of neural-enhanced Edge Intelligence system, which handles the end-to-end processing procedure by synthetically optimizing the key components of large-scale learning paradigm, hardware-adaptive neural architecture and communication-efficient cross-device interaction. As a fundamental infrastructure to bring edge intelligence to the production environment, the Edge Intelligence system offers great advantages in terms of lightweight on-device data analytics, real-time streaming processing and perfect rate-distortion balance on distributed collaborative devices. This talk gives a comprehensive inspection of implementing high-performance Edge Intelligence systems, covering the realistic design challenges, possible solutions and promising research opportunities.