【Talk】Prof. Zhengya Zhang (U Michigan): Neuromorphic Computing Using Sparse Codes: From Algorithm to Hardware (July 16, 2015(Thursday, 10:30am-12pm)
Invite you all to join it. 歡迎踴躍參加 !
Title: Neuromorphic Computing Using Sparse Codes: From Algorithm to Hardware
Date: July 16, 2015 ( Thursday, 10:30 am ~ 12:00 pm)
Place: ED528, 5F, Engineering Building 4, NCTU
交通大學(光復校區)工程四館5樓528室
Speaker: Prof. Zhengya Zhang (University of Michigan, Ann Arbor)
Abstract:
Some of the latest advances in computer vision have been built upon the understanding of the mammalian primary visual cortex (V1). The receptive fields of V1 neurons can be compared to the basis functions underlying natural images. Learning the receptive fields allows us to carry out complex vision processing, including efficient image encoding, feature detection, and classification. Sparse coding is one development in unsupervised machine learning for training a network of neurons using natural images to extract the receptive fields that resemble the V1 receptive fields. We explore the dynamics of the sparse coding algorithm for an efficient mapping onto practical hardware. Design considerations involving tuning network and neuron responses have a significant impact on the neuron spiking pattern that determines the fidelity of image processing and the efficiency of resource utilization. The spiking pattern can be further exploited to improve the performance and scalability of the hardware architecture. The soft neural computation is intrinsically error tolerant and many opportunities exist in approximating the neuron communication and computation in designing high-performance and energy-efficient image processing hardware.
Biography:
Zhengya Zhang received the B.Sc. degree from the University of Waterloo in Canada in 2003, and the M.S. and Ph.D. degrees from the University of California, Berkeley, in 2005 and 2009, respectively. Since 2009, he has been with the Department of Electrical Engineeringand Computer Science at the University of Michigan, Ann Arbor, where he is currently an Associate Professor. His research is in the area of low-power and high-performance VLSI circuits and systems for computing, communications and signal processing. Dr. Zhang received the Intel Early Career Faculty Award in 2013, the National Science Foundation CAREER Award in 2011, the David J. Sakrison Memorial Prize from UC Berkeley in 2009, and the Best Student Paper Award at the Symposium on VLSI Circuits in 2009. He is an Associate Editor of the IEEE Transactions on Circuits and Systems-I, II, and the IEEE Transactions on Very Large Scale Integration (VLSI) Systems.
Host: 交大電子系楊家驤教授 Email: chy@nctu.edu.tw
Search