Learning big Gaussian Bayesian networks: partition, estimation, and fusion
Learning big Gaussian Bayesian networks: partition, estimation, and fusion
复制标题
学习大型高斯贝叶斯网络：划分、估计和融合
DOI：
--
复制
发表时间：
2019
影响因子：
6
通讯作者：
Qing Zhou
中科院分区：
文献类型：
--
作者：
J. Gu;Qing Zhou
贝叶斯网络结构学习一直是一个具有挑战性的问题。如今，具有数千个或更多节点但样本较少的海量网络经常出现在许多领域。我们开发了一种分而治之的框架，称为划分-估计-融合(PEF)，用于此类大型网络的结构学习。该方法首先将节点划分成簇，然后在每个节点簇上学习一个子图，最后将所有学习到的子图融合到一个贝叶斯网络中。PEF方法设计灵活，使得在第二步中可以使用任何结构学习方法来学习作为DAG或CPDAG的子图结构。在聚类阶段，采用层次聚类的方法自动选择合适的聚类个数。在融合步骤中，我们提出了一种在子图之间顺序添加边的新的混合方法。大量的数值实验表明，与现有方法相比，我们的PEF方法在速度和精度方面都具有竞争力。我们的方法可以将结构学习的准确率提高20%或更多，同时将运行时间减少两个数量级。
Structure learning of Bayesian networks has always been a challenging problem. Nowadays, massive-size networks with thousands or more of nodes but fewer samples frequently appear in many areas. We develop a divide-and-conquer framework, called partition-estimation-fusion (PEF), for structure learning of such big networks. The proposed method first partitions nodes into clusters, then learns a subgraph on each cluster of nodes, and finally fuses all learned subgraphs into one Bayesian network. The PEF method is designed in a flexible way so that any structure learning method may be used in the second step to learn a subgraph structure as either a DAG or a CPDAG. In the clustering step, we adapt the hierarchical clustering method to automatically choose a proper number of clusters. In the fusion step, we propose a novel hybrid method that sequentially add edges between subgraphs. Extensive numerical experiments demonstrate the competitive performance of our PEF method, in terms of both speed and accuracy compared to existing methods. Our method can improve the accuracy of structure learning by 20% or more, while reducing running time up to two orders-of-magnitude.
影响因子：
2.7
作者：
Yiping Yuan;Xiaotong Shen;W. Pan;Zizhuo Wang
通讯作者：
Yiping Yuan;Xiaotong Shen;W. Pan;Zizhuo Wang