| 面向21世纪的计算智能研究与思考 |
| |
| 引用本文： | 董聪 郭晓华. 面向21世纪的计算智能研究与思考[J]. 计算机科学, 2000, 27(2): 1-5 |
| |
| 作者姓名： | 董聪 郭晓华 |
| |
| 作者单位： | 1. 清华大学,北京,100084
2. 中国科学院模式识别国家重点实验室,北京,100080 |
| |
| 基金项目： | 国家自然科学基金(59505011,59778039)、863计划(863-2-441)和国家重点实验室基金资助项目 |
| |
| 摘 要： | 1 引言人工智能(AI),广义地讲就是用计算机模拟和再现人类的某些智能行为。从方法论的角度,AI大致可分为3种基本类型:1)以符号操作为基本特征,以演绎方法为逻辑主体,从抽象层次模拟和再现人类某些智能行为的符号机制(Symbolism);2)以大量神经元的相互连接和作用为基本特征,以归纳方法、尤其是不完全归纳方法为逻辑主体,从神经元的连接关系和作用机制层次模拟和再现人类某些智能行为的连接机制
|
| 关 键 词： | 计算智能 人工智能 符号操作 神经元 |
Research and Thought on Computational Intelligence toward the 21st Century |
| |
| Abstract: | Some great achievements have been made in the field of artificial intelligence based on the connectionism and evolutionism in the past two decades, but a few of questions have existed,about which a survey is made in the present paper. A deep study and exploration is also made on the focus questions which have been interested by scientists in the world,the origins of these focus questions are discovered, and some solving methods are suggested. Based on above works,a trend prediction about the future development of artificial intelligence is made. |
| |
| Keywords: | Artificial intelligence Connectionism Evolutionism Generalized genetic algorithm Simulated annealing algorithm |
| 本文献已被 CNKI 维普 万方数据 等数据库收录！ |
|
点击此处可从《计算机科学》下载全文 |
|
