QQ客服 官方微博 反馈留言

	多Agent交互动态影响图的近似行为等价算法; Improved behavior equivalence algorithm of multi-agent interactive dynamic influence diagrams
	田乐 ; 罗键 ; 曹浪财
	2014-04-18
关键词	多智能体 分布式人工智能 交互式动态影响图 行为等价 策略树 信度 multi-agent distributed artificial intelligence interactive dynamic influence diagrams behavioral equivalence policy tree belief
英文摘要	结合前瞻搜索思想提出了一种判断模型近似行为等价的方法,首先通过比较候选模型的部分解(即策略树)判断模型近似行为等价,然后自上而下对近似行为等价模型进行快速聚类和修剪,利用代表模型将交互式动态影响图扩展成为平铺动态影响图,最后求解平铺动态影响图.算法减少了候选模型的存储空间和运行时间,提高了算法的效率.最后通过多AgEnT老虎问题及音乐会问题的实验验证了该方法的有效性.; The look-ahead search method was used to give a new method for determining approximate behavior equivalence.The method first determined whether the models were approximately behavior equivalent by comparing part of the solution(i.e.policy tree),then quickly clustered top-down and pruned the models that were approximately behavior equivalent.Next,the method used representative model to expand the interactive dynamic influence diagrams into flat dynamic influence diagrams.Finally,the flat dynamic influence diagrams were solved.The method reduces the storage space and the running time,thus improves the efficiency of the algorithm.The effectiveness of the proposed method was verified through experiments on multi-agent tiger and multi-agent concert problems.; 国家自然科学基金资助项目(61375070)
语种	zh_CN
内容类型	期刊论文
源URL	[http://dspace.xmu.edu.cn/handle/2288/105771]
专题	航空航天－已发表论文
推荐引用方式 GB/T 7714	田乐,罗键,曹浪财. 多Agent交互动态影响图的近似行为等价算法, Improved behavior equivalence algorithm of multi-agent interactive dynamic influence diagrams[J],2014.
APA	田乐,罗键,&曹浪财.(2014).多Agent交互动态影响图的近似行为等价算法..
MLA	田乐,et al."多Agent交互动态影响图的近似行为等价算法".(2014).

个性服务

查看访问统计

相关权益政策

暂无数据

收藏/分享

除非特别说明，本系统中所有内容都受版权保护，并保留所有权利。