论文部分内容阅读
基于“物理符号系统假设”的传统人工智能采信了低阶结构不连续的思想,将概念化与概念的语义基础分离,把对思维过程的模拟看作是能用形式化方法来实现的。但事实证明,这种后验性方法存在理论和实现上的双重危机,要完成有实际意义的、有创新性的智能行为丰富的语义是必需的,这使语义问题成为人工智能不同应用分支中的焦点,它包括语义获取、表达和使用三个方面。而要实现对语义问题的认识和解决就必须把它和其它智能行为作为一个连续的、相关的、不可分割的认知结构进行完整地考察,以系统的观点来看待智能模型的构造问题。这个认知结构统一性的基石就是基于神经生理基础的、以知觉的心理生理学解释为依据的、对语义的直接表达,这为统一以神经系统动力学为模型的其它各种智能行为提供了基础。
The traditional artificial intelligence based on “physical symbol system hypothesis ” adopted the idea of discontinuity of low-order structure, separated the conceptualization from the semantic foundation of concept, and regarded the simulation of thinking process as a formal method. However, this posterior approach proves to be a double crisis both in theory and implementation, and it is necessary to achieve practical and innovative sentiment rich in intelligence and behavior. This makes the semantic problem a branch of artificial intelligence in different applications The focus includes three aspects of semantic acquisition, expression and use. To realize the understanding and solution of the semantic problems, we must consider it and other intelligent behaviors as a continuous, relevant and indivisible cognitive structure in a holistic way and view the problem of the construction of the intelligent model systematically. The cornerstone of the unity of cognitive structure is based on the neurophysiological basis, based on perceptual psychophysiological interpretation, the direct expression of semantics, which provides the basis for the unification of various other intelligent behaviors that model neuro-system dynamics .