语境与人工智能  

Context and Artificial Intelligence

作　　者：魏屹东[1] WEI Yidong(School of Philosophy,Shanxi University,Taiyuan 030006,China)

机构地区：[1]山西大学哲学学院,山西太原030006

出　　处：《山西大学学报（哲学社会科学版）》2024年第1期43-52,共10页Journal of Shanxi University(Philosophy and Social Science Edition)

基　　金：国家社科基金重大项目“人工认知对自然认知挑战的哲学研究”(21&ZD061)。

摘　　要：使用语言的人类是语境化的,因而人类产生的一切知识无疑也是语境化的。事物和行动的语境性在哲学和语言学研究中已受到了重视,形成了关于世界假设的语境论、科学语境论和认识的语境论等。而在哲学和语言学之外,特别是在计算机科学和人工智能领域,语境的作用尚未受到足够的重视,这对于人们建构通用人工智能很不利。实际上,计算机科学和人工智能专家都自觉或不自觉地使用了语境进行建模和推理,因此,需要阐明人工智能和认知科学中语境的含义、语境中的多智能体建模和推理,以及人工主体的语境性与组合性的关系。Human beings who use language are contextualized,so all knowledge generated by humans is undoubted⁃ly contextualized.The contextuality of things and actions in context has been valued and studied in philosophy and linguistics,forming a contextualism about world hypothesis,scientific contextualism and epistemic contextualism,etc.,while outside philosophy and linguistics,especially in the fields of computer science and artificial intelli⁃gence,the role of context has not received due attention and research,which is very detrimental to people’s con⁃struction of artificial general intelligence.In fact,experts in computer science and artificial intelligence have been using context consciously or unconsciously for modeling and reasoning,so it is necessary to clarify the meaning of context in artificial intelligence and cognitive science,multi-agent’s modeling and reasoning in context,and the relations between contextuality and compositionality of artificial agents.

关 键 词：语境 语境性 人工智能 认知科学 适应性表征 

分 类 号：B81[哲学宗教—逻辑学]