arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2103.13796cs.LGstat.ML
收藏

观测环境下贝叶斯网络的主动结构学习

Active Structure Learning of Bayesian Networks in an Observational Setting

Ben-Gurion Univesity of the Negev(内盖夫本-古里安大学)

Noa Ben-David, Sivan Sabato

摘要:

我们研究了贝叶斯网络的主动结构学习在一个观察的设置,其中有外部限制的变量值的数量,可以从相同的样本。从网络变量的联合分布中抽取随机样本,并且算法迭代地选择在下一个样本中观察哪些变量。我们提出了一种新的主动学习算法,这种设置,发现一个高概率的结构与分数是$ε$ -接近最佳分数。我们表明,对于一类分布,我们长期稳定,样本复杂性降低到一个因素的$\widetildeΩ(d^3)$,可以得到,其中$d$是网络变量的数量。我们进一步表明,在最坏的情况下,样本复杂性的主动算法是保证几乎相同的一个天真的基线算法。为了补充理论结果,我们报告的实验,比较新的主动算法的性能,天真的基线,并证明样本复杂度的改善。在https://github.com/noabdavid/activeBNSL上提供了算法和实验的代码。

英文摘要:

We study active structure learning of Bayesian networks in an observational setting, in which there are external limitations on the number of variable values that can be observed from the same sample. Random samples are drawn from the joint distribution of the network variables, and the algorithm iteratively selects which variables to observe in the next sample. We propose a new active learning algorithm for this setting, that finds with a high probability a structure with a score that is $ε$-close to the optimal score. We show that for a class of distributions that we term stable, a sample complexity reduction of up to a factor of $\widetildeΩ(d^3)$ can be obtained, where $d$ is the number of network variables. We further show that in the worst case, the sample complexity of the active algorithm is guaranteed to be almost the same as that of a naive baseline algorithm. To supplement the theoretical results, we report experiments that compare the performance of the new active algorithm to the naive baseline and demonstrate the sample complexity improvements. Code for the algorithm and for the experiments is provided at https://github.com/noabdavid/activeBNSL.

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。

↑