Visualizing the effects of predictor variables in black box supervised learning models

In many supervised learning applications, understanding and visualizing the effects of the predictor variables on the predicted response is of paramount importance. A shortcoming of black box supervised learning models (e.g. complex trees, neural networks, boosted trees, random forests, nearest neig...

Full description

Saved in:
Bibliographic Details
Published in:Journal of the Royal Statistical Society. Series B, Statistical methodology Vol. 82; no. 4; pp. 1059 - 1086
Main Authors: Apley, Daniel W., Zhu, Jingyu
Format: Journal Article
Language:English
Published: Oxford Wiley 01.09.2020
Oxford University Press
Subjects:
artificial intelligence
Dependence
Extrapolation
Forests
Functional analysis of variance
Learning
Marginal plots
materials
methodology
Neural networks
Original Articles
Partial dependence plots
Regression analysis
Statistical methods
Statistics
Supervised learning
Support vector machines
Transparency
Trees
Variables
Visualization
ISSN:1369-7412, 1467-9868
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:In many supervised learning applications, understanding and visualizing the effects of the predictor variables on the predicted response is of paramount importance. A shortcoming of black box supervised learning models (e.g. complex trees, neural networks, boosted trees, random forests, nearest neighbours, local kernel-weighted methods and support vector regression) in this regard is their lack of interpretability or transparency. Partial dependence plots, which are the most popular approach for visualizing the effects of the predictors with black box supervised learning models, can produce erroneous results if the predictors are strongly correlated, because they require extrapolation of the response at predictor values that are far outside the multivariate envelope of the training data. As an alternative to partial dependence plots, we present a new visualization approach that we term accumulated local effects plots, which do not require this unreliable extrapolation with correlated predictors. Moreover, accumulated local effects plots are far less computationally expensive than partial dependence plots.We also provide an R package ALEPlot as supplementary material to implement our proposed method.
Bibliography:ObjectType-Article-1
SourceType-Scholarly Journals-1
ObjectType-Feature-2
content type line 14
content type line 23
ISSN:1369-7412
1467-9868
DOI:10.1111/rssb.12377