Visualizing the Effects of Predictor Variables in Black Box Supervised Learning Models

Abstract

When fitting black box supervised learning models (e.g., complex trees, neural networks, boosted trees, random forests, nearest neighbors, local kernel-weighted methods, etc.), visualizing the main effects of the individual predictor variables and their low-order interaction effects is often important, and partial dependence (PD) plots are the most popular approach for accomplishing this. However, PD plots involve a serious pitfall if the predictor variables are far from independent, which is quite common with large observational data sets. Namely, PD plots require extrapolation of the response at predictor values that are far outside the multivariate envelope of the training data, which can render the PD plots unreliable. Although marginal plots (M plots) do not require such extrapolation, they produce substantially biased and misleading results when the predictors are dependent, analogous to the omitted variable bias in regression. We present a new visualization approach that we term accumulated local effects (ALE) plots, which inherits the desirable characteristics of PD and M plots, without inheriting their preceding shortcomings. Like M plots, ALE plots do not require extrapolation; and like PD plots, they are not biased by the omitted variable phenomenon. Moreover, ALE plots are far less computationally expensive than PD plots.

arXiv commentThe R package ALEPlot is available on CRAN. The new version contains refined definitions of ALE effects, a new illustrative example, theorems and proofs of asymptotic properties of ALE effects and estimators, and extra implementation details

Cited by in corpus (23)

• The role of explainability in creating trustworthy artificial intelligence for health care: a comprehensive survey of the terminology, design choices, and evaluation strategies
• Interpretable Machine Learning -- A Brief History, State-of-the-Art and Challenges
• Statistical stability indices for LIME: obtaining reliable explanations for Machine Learning models
• Machine learning algorithms to infer trait-matching and predict species interactions in ecological networks
• Model-agnostic Feature Importance and Effects with Dependent Features -- A Conditional Subgroup Approach
• Explanations of model predictions with live and breakDown packages
• Explainable Artificial Intelligence for Process Mining: A General Overview and Application of a Novel Local Explanation Approach for Predictive Process Monitoring
• A Survey of Explainable Artificial Intelligence (XAI) in Financial Time Series Forecasting
• Quantifying Model Complexity via Functional Decomposition for Better Post-Hoc Interpretability
• Sampling, Intervention, Prediction, Aggregation: A Generalized Framework for Model-Agnostic Interpretations
• A Machine Learning Tutorial for Operational Meteorology, Part I: Traditional Machine Learning
• OptiLIME: Optimized LIME Explanations for Diagnostic Computer Algorithms
• Walling up Backdoors in Intrusion Detection Systems
• A new paradigm for medium-range severe weather forecasts: probabilistic random forest-based predictions
• Modeling Heterogeneity in Mode-Switching Behavior Under a Mobility-on-Demand Transit System: An Interpretable Machine Learning Approach
• In Search of netUnicorn: A Data-Collection Platform to Develop Generalizable ML Models for Network Security Problems
• Towards Explainability of Machine Learning Models in Insurance Pricing
• Enhancing Explainability of Neural Networks through Architecture Constraints
• Understanding Consumer Preferences for Explanations Generated by XAI Algorithms
• Better Model Selection with a new Definition of Feature Importance
• A Physics-Constrained Neural Differential Equation Framework for Data-Driven Snowpack Simulation
• Technical Report: Partial Dependence through Stratification
• Interpretabilité des modèles : état des lieux des méthodes et application à l'assurance