神经计算科学的展望

Prospects for Neurocomputing Science

神经计算科学是从信息科学的角度来研究如何加速神经网络模仿和延伸人脑的高级精神活动 ,如联想、记忆、推理、思维及意识等智能行为。这涉及到脑科学、认知科学 ,神经生物学、非线性科学、计算机科学、数学、物理学诸学科的综合集成。它是综合研究和实现类脑智能信息系统的一个新思想和新策略。本文仅重点论述神经智能信息处理、盲信号处理技术、演化认知与进化计算、混沌信息技术及软科学的开发 ,由此展望神经计算科学的未来发展及潜力 ,可带动诸学科及信息技术的美好发展。

Neurocomputing science is a high level mind activities, which researches on how to speed up neural network to imitate and extend the functions of human brain as viewed from information science, such as association, memory, inference, thinking and consciousness intelligent behaviors and so on. This science involves synthetical integration of the various branches of learning like brain science, cognition science, neural biology, nonlinear science, computer science, mathematics and physics. Neurocomputing science is a new thought and a new strategy, which synthetically studies and realizes brainlike intelligent information system. In the paper, it is emphatically discussed for the development of the neural intelligent information processing, blind signal processing technique, evolving cognition and evolution computation, and the chaotic information technique and soft science. Thus, having been looked forward to the future development and potentiality, the neurocomputing science will drive the glorious development of the various branches of learning and information techniques(IT).