论文部分内容阅读
神经计算是一门涉及神经网络的设计实现及应用的新的工程学科。神经网络是非算法的计算结构,其拓扑是有向图,它可以通过对连续或初始输入的状态响应来进行信息处理。神经网络中的节点称为处理单元,有向边(信息通道)称为互连通路。神经网络已经表明能够进行如下的信息处理操作:感觉信息处理(模式预处理,模式识别),知识处理(知识表示,从数据中自动抽取知识,对不精确及矛盾的知识进行推理)和控制(平滑,快速机器人手腿控制,机器人“手眼协调”)。
Neural computing is a new engineering discipline that involves the design and application of neural networks. Neural networks are non-algorithmic computational structures whose topologies are directed graphs that can process information by responding to the state of successive or initial inputs. Nodes in the neural network are called processing units and directed edges (information channels) are called interconnecting paths. Neural networks have demonstrated the ability to perform the following information manipulation operations: sensory information processing (pattern preprocessing, pattern recognition), knowledge processing (knowledge representation, automatic knowledge extraction from data, reasoning about inaccurate and contradictory knowledge) and control Smooth, fast robot leg control, robot “hand-eye coordination”).