Automated Dependence Plots

TL;DR

This paper introduces Automated Dependence Plots (ADP), a method that automates the selection of informative plots and extends PDPs to arbitrary feature directions, enhancing model interpretability and validation.

Contribution

It formalizes an automated approach for selecting interesting dependence plots and extends PDPs beyond single features to arbitrary directions, improving model analysis tools.

Findings

ADP effectively automates the identification of key dependence plots.

ADP extends PDPs to show responses along arbitrary feature directions.

Demonstrated usefulness in model selection, bias detection, and latent space exploration.

Abstract

In practical applications of machine learning, it is necessary to look beyond standard metrics such as test accuracy in order to validate various qualitative properties of a model. Partial dependence plots (PDP), including instance-specific PDPs (i.e., ICE plots), have been widely used as a visual tool to understand or validate a model. Yet, current PDPs suffer from two main drawbacks: (1) a user must manually sort or select interesting plots, and (2) PDPs are usually limited to plots along a single feature. To address these drawbacks, we formalize a method for automating the selection of interesting PDPs and extend PDPs beyond showing single features to show the model response along arbitrary directions, for example in raw feature space or a latent space arising from some generative model. We demonstrate the usefulness of our automated dependence plots (ADP) across multiple use-cases and datasets including model selection, bias detection, understanding out-of-sample behavior, and exploring the latent space of a generative model.