论文部分内容阅读
传统的BDI审慎式智能体一直缺乏一种高效的实现机制,而且不具备通过利用由应用环境反馈的信息来完善其性能的能力。为此,在传统智能体系统操作模式基础上引入有效的学习和感知机制,提出了一个基于事例推理(CBR)的自适应审慎智能体系统结构模型。该模型通过建立CBR系统和智能体认知理论之间的映射关系来构造审慎智能体,并利用CBR的认知-学习循环周期不断完善智能体的知识库来提高其在动态环境下求解问题的自适应性,为用户提供更好的求解效果。
The traditional BDI prudential agents have always lacked an efficient implementation mechanism and did not have the capability to improve their performance by leveraging the information fed back by the application environment. Therefore, based on the traditional operation mode of intelligent agent system, an effective learning and perception mechanism is introduced, and a case-based reasoning (CBR) adaptive prudent agent system structural model is proposed. The model constructs the prudential agent by establishing the mapping relationship between the CBR system and the cognitive theory of the agent and uses the CBR cognitive-learning cycle to continuously improve the knowledge base of the agent to improve its solving problem under the dynamic environment Adaptive, to provide users with a better solution.