面向计数数据的有向无环图(DAG)引导结构学习
Guided structure learning of DAGs for count data
AI总结:
针对计数数据的DAG结构学习问题,提出一种利用变量拓扑排序先验的新算法,证明其渐近一致性,并通过有限样本实验与主流方法比较性能。
摘要:
本文研究有向无环图(DAG)的结构学习,核心思想是利用当前领域可获得的先验知识来引导对最佳结构的搜索。具体而言,除给定数据外,我们假设已知变量的拓扑排序。我们研究了一种用于学习DAG结构的新算法,并证明其在观测数趋于无穷的极限下具有理论一致性。此外,我们通过实验将所提算法与多种流行的竞争算法进行比较,以研究其在有限样本下的表现。
英文摘要:
In this paper, we tackle structure learning of Directed Acyclic Graphs (DAGs), with the idea of exploiting available prior knowledge of the domain at hand to guide the search of the best structure. In particular, we assume to know the topological ordering of variables in addition to the given data. We study a new algorithm for learning the structure of DAGs, proving its theoretical consistence in the limit of infinite observations. Furthermore, we experimentally compare the proposed algorithm to a number of popular competitors, in order to study its behavior in finite samples.
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。