脑与认知技术发展综述

脑与认知是利用神经形态计算来模拟生物大脑处理信息的过程,被认为是探究甚至实现人工智能的最直接技术路线,而且可以为人工神经网络研究提供基础导向,因为人类和动物"先天"的结构对其技能和习性有着决定性作用。对近期脑与认知的相关研究与成果进行了综合评述,并对未来发展趋势进行了展望。综述表明,类脑仿生研究与应用改进了人工智能系统,重建复杂的大脑皮层神经网与回路是当前研究重点;脉冲神经网络被认为是目前深度神经网络的"继承者";脑机接口技术发展呈现非侵入式趋势,面向语言障碍病患实现语音合成的脑机接口技术成为近期新兴研究方向。

A Survey on Brain and Cognition Technology

Brain and cognition aims at processing information by leveraging neural morphological computation.It is regarded as the most direct technical route to explore and realize artificial intelligence(AI),and meanwhile it could provide basic guidance for the research of artificial neural network,because the inborn structure of human and animal plays a decisive role in their skills and habits generation.This paper reviews the recent research and achievements of brain and cognition technology,and looks forward to the future development trend.It mainly includes three aspects:the research and application of bionics has improved the AI system,and the reconstruction of complex neural networks and circuits in the cerebral cortex is the focus of current research;pulse neural networks are considered as the inheritors for current deep neural networks;the development of brain computer interface technology presents a non-invasive trend,and speech synthesis for patients with language disorders has become a new research direction in recent years.