Best-First vs. Depth-First AND/OR Search for Multi-objective Constraint Optimization

Best-First vs. Depth-First AND/OR Search for Multi-objective Constraint Optimization
复制标题

多目标约束优化的最佳优先与深度优先 AND/OR 搜索

DOI:
10.1109/ictai.2010.69
复制
发表时间:
2010
期刊:
2010 22nd IEEE International Conference on Tools with Artificial Intelligence
影响因子:
--
通讯作者:
Radu Marinescu
Radu Marinescu
中科院分区:
--
文献类型:
--
作者:
Radu Marinescu

文献摘要

参考文献

被引文献

相似文献

在本文中,我们提出和评估的权力最好的第一次搜索与/或搜索空间的多目标约束优化。搜索空间的AND/OR表示的主要优点是它对问题结构的敏感性,这可以转化为显着的时间节省。我们引入了一个线性空间最佳优先搜索算法,探索了一个与/或搜索树,并使用一类基于分区的算法的指导。在多目标约束优化的随机和现实世界基准测试中,经验证明了最佳优先方法优于使用相同启发式函数的深度优先AND/OR分支定界搜索。
In this paper we present and evaluate the power of best-first search over AND/OR search spaces for multi-objective constraint optimization. The main virtue of the AND/OR representation of the search space is its sensitivity to problem structure, which can translate into significant time savings. We introduce a linear-space best-first search algorithm that explores an AND/OR search tree and uses a class of partitioning-based heuristics for guidance. The superiority of the best-first approach over depth-first AND/OR Branch-and-Bound search using the same heuristic function is demonstrated empirically on random and real-world benchmarks for multi-objective constraint optimization.
J.Appl.Phys.57-(1)。
DOI: --
发表时间: --
期刊:
影响因子: --
作者:
通讯作者: --