人工智能的适应性表征

Adaptive Representation of Artificial Intelligence

人类发展史和进化论业已表明,人类是环境适应性的,进而人的认知也是环境适应性的,这样一来,人类创造的知识就是一种适应性表征。按照这种逻辑和思路,作为人造物的人工智能的表征也应该是适应性的。基于这种想法,自然就会形成这样的问题:什么是人工智能的适应性表征,适应性表征的主体是什么,其表征方法论是什么,它有哪些表征方式和类型等。这些问题构成了人工智能的适应性表征的问题域,通过对这些问题的探索性解答,说明人工智能是理性主体,其表征也必然是基于人类理性认知的适应性表征,同样遵循人类的认知模式,即理性地行动。

The development of human history and the theory of evolution have shown that man is environmentally a- daptable, and human cognition is also environmentally adaptable. In this way, knowledge made by man is a kind of adap- tive representation. Following this logic or way of thinking, the representation in artificial intelligence as an artifact should also be adaptable. Then, what is an adaptive representation of artificial intelligence？ What is the agent of adaptive repre- sentation？ What is the methodology of adaptive representation？ How many types and ways of adaptive representation are there？ These questions form the problem domain of adaptive representation. By giving some tentative answers to these questions, the paper tries to show that artificial intelligence has its rational agent, and its representation must be adaptive representation based on human rationality and also follows the cognitive model of human beings, namely, it acts rational- ly.