摘要
针对模式表示研究存在的语义缺失问题,基于agent技术和人的记忆原理,提出一个新的模式表示模型agent影响图(agent influence map,AIM)。AIM反映了模式的整体特征,提供一个有效的软计算工具来支持基于先验知识的自适应行为。AIM通过特征的多阶段整合呈现记忆模式的层次性;把模式信息存储在整个网络中,通过协作涌现出高层次特征体现记忆的语义特性。
To solve the existing problems of semanticloss in pattern expression,a new model of pattern expression was proposed,which is called the AIM（agent influence map）that is based on an agent and memory principle.AIM reflected the whole character of the pattern,and it provided an effective soft computing tool to support adaptive behavior that is based on prior knowledge.AIM showed the hierarchy of the memory model by integrating the character in multi-stages,while the pattern information is stored in the entire network,and high-level features are manifested by collaboration,and demonstrate the memory＇s semantic features.
出处
《山东大学学报（工学版）》
CAS
北大核心
2010年第5期72-76,共5页
Journal of Shandong University（Engineering Science）
基金
国家自然科学基金资助项目(60873069)
江苏省研究生创新项目(CX99B204)
关键词
智能体
模式表示
定性特征
agent
pattern expression
qualitative character
