Abstract

Variable importance, interaction measures, and partial dependence plots are important summaries in the interpretation of statistical and machine learning models. In this article, we describe new visualization techniques for exploring these model summaries. We construct heatmap and graph-based displays showing variable importance and interaction jointly, which are carefully designed to highlight important aspects of the fit. We describe a new matrix-type layout showing all single and bivariate partial dependence plots, and an alternative layout based on graph Eulerians focusing on key subsets. Our new visualizations are model-agnostic and are applicable to regression and classification supervised learning settings. They enhance interpretation even in situations where the number of variables is large. Our R package vivid (variable importance and variable interaction displays) provides an implementation. Supplementary files for this article are available online.

摘要

变量重要性、交互作用度量和部分依赖图是解释统计和机器学习模型的重要总结。在本文中，我们描述了用于探索这些模型摘要的新可视化技术。我们构建了热图和基于图形的显示，共同显示变量的重要性和交互，这些显示经过精心设计，以突出拟合的重要方面。我们描述了一种新的矩阵类型布局，显示所有单变量和二元部分依赖图，以及基于图欧拉算法的替代布局，重点关注关键子集。我们的新可视化与模型无关，适用于回归和分类监督学习设置。即使在变量数量很大的情况下，它们也能增强解释。我们的 R 包vivid（可变重要性和可变交互显示）提供了一个实现。本文的补充文件可在线获取。