语境论视野下的人工智能范式发展趋势研究  

Research on the Paradigm Development Trend of Artificial Intelligence in View of Contextualism

作　　者：董佳蓉[1] 

机构地区：[1]山西大学哲学社会学学院,太原030006

出　　处：《科学技术哲学研究》2011年第2期33-38,共6页Studies in Philosophy of Science and Technology

基　　金：国家社会科学基金青年项目"人工智能语境论范式研究"(10CZX014)

摘　　要：人工智能在短短50多年时间里出现了符号主义、连接主义和行为主义等三种主导性范式,并表现出一定的融合趋势。文章认为,独特的表征理论和计算理论是人工智能范式得以成立的基础。文章从表征和计算两个角度对人工智能现有范式进行分析,认为三者面临的共同问题,都是难以在不断变化着的真实语境中很好地处理问题。人工智能将紧紧围绕语境问题展开研究。因此,人工智能研究范式也必然不会停留在现有范式之内,需要构建语境论视野下的范式发展观。In a short period of over 50 Years,there appeared three kinds of predominant paradigms of artificial intelligence which are symbolism,connectionism and behaviorism and a certain fusion trend has come out.This article thinks that unique representation theory and computation theory are the foundation of artificial intelligence.This article analyses the existing paradigms from representation and computation,and thinks that the common problem of the above three paradigms is that it is hard for then to handle problems in real contexts which is continuously changing.Research in artificial intelligence will be closely related to context problems.Therefove it is necessary to set up the paradigm development view in view of contextualism.

关 键 词：人工智能范式 表征 计算 

分 类 号：N02[自然科学总论—科学技术哲学]