摘要
根据神经行为学和认知心理学观点提出智能控制系统的新构型 ,从相对熵最小化原理构建智能控制理论 .从神经生理学和神经网络理论证明分层智能控制系统的基本原理IPDI(Increasingprecisionwithdecreasingintelli gence) [1,2 ] .从相对熵最小化观点和IPDI原理出发 ,讨论智能系统的智能行为 ,给出关于智能系统学习过程的几个定理和学习算法 .
Based on the point of view of neuroethology and cognition psychology,a new configuration of intelligent control system is presented in this paper; the theory of intelligent control is constructed by principle of minimizing of relatvie entropy.The basic principle of hierarchical intelligent control (IPDI_Increasing precision with decreasing intelligence)is proved from points of view of the neurophysiology and the theory of neural network.Intelligent behavior of intelligent system is discussed from the viewpoint of minimizing of relative entropy and from the principle of IPDI;several theorems and algorithms for the learning process of intelligent system are also given in this paper.
出处
《控制理论与应用》
EI
CAS
CSCD
北大核心
1999年第1期27-31,共5页
Control Theory & Applications
基金
国家自然科学基金!( 6 96 73 0 0 2 )
关键词
智能控制
相对熵
IPDI原理
学习功能
记忆功能
intelligent control
relative entropy
principle of IPDI
function of learning
function of associative memory
function of induction
