Born-Again Tree Ensembles

Born-Again Tree Ensembles
复制标题

重生树合奏

DOI:
--
复制
发表时间:
2020
期刊:
International Conference on Machine Learning
影响因子:
--
通讯作者:
Maximilian Schiffer
Maximilian Schiffer
中科院分区:
--
文献类型:
--
作者:
Thibaut Vidal;Toni Pacheco;Maximilian Schiffer

文献摘要

参考文献

被引文献

相似文献

机器学习算法在金融、医学和刑事司法中的应用可以深刻地影响人类的生活。因此,对可解释机器学习的研究迅速发展,试图更好地控制和修复可能的错误和偏见来源。树集成在不同的领域提供了良好的预测质量,但同时使用多棵树降低了集成的可解释性。在此背景下,我们研究重生树集成,即构建最小大小的单个决策树的过程,该决策树在其整个特征空间中再现与给定树集成完全相同的行为。为了找到这样的树,我们开发了一种基于动态规划的算法,该算法利用复杂的修剪和边界规则来减少递归调用的数量。该算法为许多实际感兴趣的数据集生成最佳再生树,导致分类器通常更简单,更易于解释,而没有任何其他形式的妥协。
The use of machine learning algorithms in finance, medicine, and criminal justice can deeply impact human lives. As a consequence, research into interpretable machine learning has rapidly grown in an attempt to better control and fix possible sources of mistakes and biases. Tree ensembles offer a good prediction quality in various domains, but the concurrent use of multiple trees reduces the interpretability of the ensemble. Against this background, we study born-again tree ensembles, i.e., the process of constructing a single decision tree of minimum size that reproduces the exact same behavior as a given tree ensemble in its entire feature space. To find such a tree, we develop a dynamic-programming based algorithm that exploits sophisticated pruning and bounding rules to reduce the number of recursive calls. This algorithm generates optimal born-again trees for many datasets of practical interest, leading to classifiers which are typically simpler and more interpretable without any other form of compromise.
DOI: 10.4310/sii.2009.v2.n3.a11
发表时间: 2009-01-01
影响因子: 0.8
作者:
Zhang H;Wang M
通讯作者: Wang M
DOI: 10.1145/3412815.3416893
发表时间: 2016-11
期刊: Proceedings of the 2020 ACM-IMS on Foundations of Data Science Conference
影响因子: --
作者:
S. Tan;Matvey Soloviev;G. Hooker;M. Wells
通讯作者: S. Tan;Matvey Soloviev;G. Hooker;M. Wells
DOI: 10.1073/pnas.1900654116
发表时间: 2019-10-29
影响因子: 11.1
作者:
Murdoch, W. James;Singh, Chandan;Yu, Bin
通讯作者: Yu, Bin