arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 2006.14937cs.LGcs.AIstat.ML
收藏

随机森林中的节理

Joints in Random Forests

Eindhoven University of Technology(埃因霍温理工大学)

Alvaro H. C. Correia, Robert Peharz, Cassio de Campos

摘要:

决策树(DT)和随机森林(RF)是强大的判别学习器,对日常机器学习实践者和数据科学家来说至关重要。然而,由于它们的区别性,它们缺乏处理具有缺失特征的输入或检测离群值的原则性方法,这需要将它们与插补技术或单独的生成模型配对。在本文中,我们通过与概率电路(一类重要的易处理概率模型)建立联系,证明DT和RF可以自然地被解释为生成模型。这种重新解释使它们在特征空间上具有完整的联合分布,并导致生成决策树(GeDT)和生成森林(GeF),这是一种新型的混合生成判别模型。该系列模型保留了DT和RF的整体特征,同时还能够通过边缘化处理缺失的特征。在某些假设下,经常贝叶斯一致性结果,我们表明,在GeDT和GeFs的一致性扩展到任何模式的输入功能缺失,如果随机缺失。从经验上讲,我们表明,我们的模型在处理缺失数据(如K-最近邻插补)方面往往优于常见的例程,此外,我们的模型可以通过监测输入特征的边际概率来自然检测离群值。

英文摘要:

Decision Trees (DTs) and Random Forests (RFs) are powerful discriminative learners and tools of central importance to the everyday machine learning practitioner and data scientist. Due to their discriminative nature, however, they lack principled methods to process inputs with missing features or to detect outliers, which requires pairing them with imputation techniques or a separate generative model. In this paper, we demonstrate that DTs and RFs can naturally be interpreted as generative models, by drawing a connection to Probabilistic Circuits, a prominent class of tractable probabilistic models. This reinterpretation equips them with a full joint distribution over the feature space and leads to Generative Decision Trees (GeDTs) and Generative Forests (GeFs), a family of novel hybrid generative-discriminative models. This family of models retains the overall characteristics of DTs and RFs while additionally being able to handle missing features by means of marginalisation. Under certain assumptions, frequently made for Bayes consistency results, we show that consistency in GeDTs and GeFs extend to any pattern of missing input features, if missing at random. Empirically, we show that our models often outperform common routines to treat missing data, such as K-nearest neighbour imputation, and moreover, that our models can naturally detect outliers by monitoring the marginal probability of input features.

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。

↑