.png){kind=icon}
返回
Dimensionality reduction in multiobjective shortest path search
.png){kind=icon}
PRE
AI
DOI：10.1016/j.cor.2015.05.007.png)
.png){kind=icon}
原文求助.png){kind=icon}
分享.png){kind=icon}
收藏
摘要
En 中文
One-to-one multiobjective search in graphs deals with the problem of finding all Pareto-optimal solution paths between given start and goal nodes according to a number of distinct noncommensurate objectives. The problem is inherently more complex than single objective graph search. Time requirements are dominated by the facts that (a) many different non-dominated labels may need to be explored for each node; (b) each new label under consideration must be checked for dominance against various subsets of previously generated labels. This paper describes how a dimensionality reduction technique can be applied to exact label-setting algorithms, reducing the number of dominance checks and allowing for much faster multiobjective search. The technique is applied to NAMOA*, a state of the art exact label-setting multiobjective search algorithm, achieving reductions in time requirements of more than an order of magnitude over problems in random grids and realistic road maps. Tests include problems with three, four, and five objectives. (C) 2015 Elsevier Ltd. All rights reserved.
Keyword:
Combinatorial optimization
Multiobjective shortest path problem
Exact label-setting algorithms
Lower bounds
AI总结
对已上传原文的论文进行重点信息的提取，主要内容包括：简要概述、研究摘要、背景介绍、关键亮点、图文解析、展望与总结。
期刊
C
IF:
4.3
论文数:
6.5K
被引数:
1.8W
机构
引用论文
.png){kind=icon}
Does it take older adults longer than younger adults to perceptually segregate a speech target from a background masker?在感知上将语音目标与背景掩蔽器隔离开来是否需要老年人比年轻人更长的时间？
PREAI
分享收藏
分享收藏
A comparison of heuristic best-first algorithms for bicriterion shortest path problems双准则最短路径问题的启发式最佳优先算法的比较
PREAI
分享收藏
查看更多内容