arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2012.07320cs.LGcs.AI
收藏

通过昂贵的评估优化离散空间:学习搜索框架

Optimizing Discrete Spaces via Expensive Evaluations: A Learning to Search Framework

Washington State University(华盛顿州立大学);Oregon State University(俄勒冈州立大学)

Aryan Deshwal, Syrine Belakaria, Janardhan Rao Doppa, Alan Fern

摘要:

我们考虑在离散空间上优化昂贵的黑盒函数的问题(例如,集合、序列、图)。关键的挑战是选择一系列组合结构进行评估,以便尽快识别高性能结构。我们的主要贡献是介绍和评估一个新的学习搜索框架,这个问题称为L2 S-DISCO。关键的洞察力是采用控制知识在每一步的搜索程序,以选择下一个结构,并改善控制知识,新的功能评估观察。我们提供了一个具体的L2 S-DISCO本地搜索程序的实例,并在不同的现实世界的基准经验评估。结果表明,在解决复杂的优化问题的L2 S-DISCO国家的最先进的算法的功效。

英文摘要:

We consider the problem of optimizing expensive black-box functions over discrete spaces (e.g., sets, sequences, graphs). The key challenge is to select a sequence of combinatorial structures to evaluate, in order to identify high-performing structures as quickly as possible. Our main contribution is to introduce and evaluate a new learning-to-search framework for this problem called L2S-DISCO. The key insight is to employ search procedures guided by control knowledge at each step to select the next structure and to improve the control knowledge as new function evaluations are observed. We provide a concrete instantiation of L2S-DISCO for local search procedure and empirically evaluate it on diverse real-world benchmarks. Results show the efficacy of L2S-DISCO over state-of-the-art algorithms in solving complex optimization problems.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑