IEEE CEDA Distinguish Lecture
题目：Driving distributed edge intelligence in camera systems
报告人：Prof. Vijaykrishnan Narayanan, Pennsylvania State University
As the computing power of end-point devices grows, there has been interest in developing distributed deep neural networks specifically for hierarchical inference deployments on multi-sensor systems. However, as the existing approaches rely on latent parameters trained by machine learning, it is difficult to preemptively select
front-end deep features across sensors, or understand individual feature’s relative importance for systematic global inference. In this talk, I will introduce a multi-view convolutional neural networks exploiting likelihood estimation that
can decrease an endpoint’s communication and energy costs by a factor of 3×, while achieving high accuracy
comparable to the original aggregation approaches. I will also show case approaches that dynamically and statically adapt the inference engine for power-efficient edge operations.
Vijay Narayanan is a Robert Noll Chair Professor of Computer Science and Engineering and Electrical Engineering at The Pennsylvania State University. He is the PI of the NSF Expeditions-in-Computing Program on Visual Cortex on Silicon and a thrust leader for the JUMP Center on Brain-Inspired Computing. He has published more than 400 papers and won several awards in recognition of his research in power-aware systems, embedded systems and computer architecture. He is a fellow of IEEE and ACM.