arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2408.11530cs.LOcs.AI
收藏

使用约束优化的可扩展知识重构

Scalable Knowledge Refactoring using Constrained Optimisation

University of Oxford(牛津大学);Czech Academy of Sciences Institute of Computer Science (CAS ICS)(捷克科学院计算机科学研究所)

Minghao Liu, David M. Cerna, Filipe Gouveia, Andrew Cropper

AI总结:

针对知识重构难以扩展的问题,提出基于文字决策变量和线性发明规则的约束优化方法,实现更快且压缩率更高的程序重构。

摘要:

知识重构通过引入新规则来压缩逻辑程序。当前方法难以扩展到大型程序。为克服这一限制,我们提出了一种约束优化重构方法。我们的第一个关键思想是基于文字而非规则来编码决策变量问题。第二个关键思想是专注于线性发明规则。我们在多个领域的实证结果表明,与之前最先进的方法相比,我们的方法可以更快地重构程序并获得更多压缩,有时压缩率提高60%。

英文摘要:

Knowledge refactoring compresses a logic program by introducing new rules. Current approaches struggle to scale to large programs. To overcome this limitation, we introduce a constrained optimisation refactoring approach. Our first key idea is to encode the problem with decision variables based on literals rather than rules. Our second key idea is to focus on linear invented rules. Our empirical results on multiple domains show that our approach can refactor programs quicker and with more compression than the previous state-of-the-art approach, sometimes by 60%.

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。

↑