Neuromorphic Brain-Inspired Computing with Hybrid Neural Networks

Abstract

Neuromorphic brain-inspired computing is believed to solve the bottleneck of traditional Von Neumann architecture computers and may promote the development of the next-generation of high-performance computer architectures. Therefore, in recent years, brain-inspired computing has received extensive attention. Some large-scale brain-inspired research projects have yielded some results, such as further increasing computer capabilities in data processing and machine learning. How there is a lack of widely used artificial neural network based on computer science and neuroscience-inspired models and algorithms. This low compatibility between software and hardware reduces the computational programming efficiency and becomes an obstacle to the development of brain-inspired computing. This paper introduces the concept of neuromorphic brain-inspired computing and then introduces the research results of Neurogrid, Spinnaker, TrueNorth, Loihi and Tianjic, respectively. Finally, the paper shares the views on the future development trend of the field of neuromorphic brain-inspired computing.