Tree Space Prototypes: Another Look at Making Tree Ensembles Interpretable
Tree Space Prototypes: Another Look at Making Tree Ensembles Interpretable
复制标题
DOI：
10.1145/3412815.3416893
复制
发表时间：
2016-11
期刊：
影响因子：
--
通讯作者：
S. Tan;Matvey Soloviev;G. Hooker;M. Wells
中科院分区：
文献类型：
--
作者：
S. Tan;Matvey Soloviev;G. Hooker;M. Wells
决策树集成在许多问题上表现良好，但无法解释。与现有的注重解释特征和预测之间关系的可解释性方法不同，我们提出了一种通过为每一类-原型浮出代表点来解释树集成分类器的替代方法。我们为梯度增强树模型引入了一种新的距离，并提出了一种新的、自适应的原型选择方法，该方法具有理论上的保证，可以灵活地在每类中选择不同数量的原型。我们在随机森林和梯度增强树上演示了我们的方法，表明当用作最近原型分类器时，原型可以和原始树集成一样好，甚至更好。在一项用户研究中，人类在使用原型时预测树集成分类器的输出方面比使用Shapley值时更好，Shapley值是一种流行的特征归因方法。因此，原型为基于特征的树集合解释提供了一种可行的替代方案。
Ensembles of decision trees perform well on many problems, but are not interpretable. In contrast to existing approaches in interpretability that focus on explaining relationships between features and predictions, we propose an alternative approach to interpret tree ensemble classifiers by surfacing representative points for each class -- prototypes. We introduce a new distance for Gradient Boosted Tree models, and propose new, adaptive prototype selection methods with theoretical guarantees, with the flexibility to choose a different number of prototypes in each class. We demonstrate our methods on random forests and gradient boosted trees, showing that the prototypes can perform as well as or even better than the original tree ensemble when used as a nearest-prototype classifier. In a user study, humans were better at predicting the output of a tree ensemble classifier when using prototypes than when using Shapley values, a popular feature attribution method. Hence, prototypes present a viable alternative to feature-based explanations for tree ensembles.