Huiru Liu, Liping Hu

(College of Artificial Intelligence, University of Chinese Academy of Sciences)

Abstract：

Simulating the brain is key to reproducing the human mind in computing systems. At present, although the progress in brain and cognitive science is not sufficient to support precise computational simulations of the entire human brain and its connections, inspired by the neural and cognitive mechanisms of the brain, computational modeling of the brain at the micro, meso, and macro levels is still of great significance for advancing scientific understanding of the mechanisms of the brain and promoting brain like artificial intelligence. Building neural network models that combine biological rationality and computational efficiency is an important challenge for cognitive brain computational simulation and brain like intelligence research. This article will start from the perspective of reviewing the historical context and development process of cognitive architecture, computational neuroscience, and neuromorphic artificial intelligence. It will introduce the global layout and progress of large-scale brain simulation, as well as the research on neuromorphic pulse neural network platforms. Finally, it will look forward to the future development direction of brain simulation and neuromorphic intelligence research.

Key Words：

cognitive architecture; computational neuroscience; brain like artificial intelligence; cross scale brain simulation; neuromorphic impulse neural network