Physics & Astronomy Seminar: Spiking Neural Network and Neuromorphic Computing Paradigm
Kun Yue, PhD candidate of the Ming Hsieh Department of Electrical and Computer Engineering, University of Southern California, will present.The spiking neural network (SNN) mimicking the operation mechanism of neurons in the human brain is a promising method to tackle the grand challenge posed by the National Academy of Engineering of U.S.,“reverse-engineering the brain”. However, traditional computing platforms including CPU and GPU are inefficient SNN implementation in terms of power consumption and time complexity. Hence, neuromorphic computing paradigm was proposed to achieve a biomimetic deployment of SNN. In this talk, I will first briefly introduce the spiking neural network (SNN). I will then discuss the promises and challenges of neuromorphic computing platforms design including digital and analog implementations. Last, I will discuss a few examples using novel nanoelectronic devices to design synapse and axon hillock circuits.
Tuesday, February 25 at 11:30am to 12:30pm
Ho Science Center, 101
1819 Oak Dr, Hamilton, NY 13346, USA