arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 1301.3856cs.LGcs.AIstat.ML
收藏

网络结构的贝叶斯分析

Being Bayesian about Network Structure

Hebrew University(希伯来大学);Stanford University(斯坦福大学)

Nir Friedman, Daphne Koller

摘要:

在许多领域,我们对分析底层分布的结构感兴趣,例如,一个变量是否是另一个变量的直接父变量。贝叶斯模型选择试图找到MAP模型并使用其结构来回答这些问题。然而,当可用数据量适中时,可能有许多模型具有不可忽略的后验。因此,我们想要计算特征的贝叶斯后验,即,包含它的所有模型的总后验概率。在本文中,我们提出了一种新的方法来完成这项任务。我们首先展示了如何有效地计算一个指数网络的总和,这些网络与网络变量的固定顺序一致。这使我们能够计算,对于给定的顺序,数据的边际概率和后验特征。然后,我们使用这个结果作为一个算法的基础,近似贝叶斯后验的功能。我们的方法使用马尔可夫链蒙特卡罗(MCMC)方法,但在排序,而不是在网络结构。有序的空间比结构的空间小得多,也更规则,并且有一个更平滑的后“空间”。我们在合成和现实生活中的数据集上提出了实证结果,将我们的方法与完整模型平均(如果可能),网络结构上的MCMC以及非贝叶斯自助方法进行了比较。

英文摘要:

In many domains, we are interested in analyzing the structure of the underlying distribution, e.g., whether one variable is a direct parent of the other. Bayesian model-selection attempts to find the MAP model and use its structure to answer these questions. However, when the amount of available data is modest, there might be many models that have non-negligible posterior. Thus, we want compute the Bayesian posterior of a feature, i.e., the total posterior probability of all models that contain it. In this paper, we propose a new approach for this task. We first show how to efficiently compute a sum over the exponential number of networks that are consistent with a fixed ordering over network variables. This allows us to compute, for a given ordering, both the marginal probability of the data and the posterior of a feature. We then use this result as the basis for an algorithm that approximates the Bayesian posterior of a feature. Our approach uses a Markov Chain Monte Carlo (MCMC) method, but over orderings rather than over network structures. The space of orderings is much smaller and more regular than the space of structures, and has a smoother posterior `landscape'. We present empirical results on synthetic and real-life datasets that compare our approach to full model averaging (when possible), to MCMC over network structures, and to a non-Bayesian bootstrap approach.

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。

↑