Conditional Feature Importance with Generative Modeling Using Adversarial Random Forests

TL;DR

This paper introduces cARFi, a novel method leveraging adversarial random forests to accurately measure conditional feature importance in tabular data, addressing challenges in generating on-manifold feature values for explainability.

Contribution

The paper presents cARFi, a new approach that uses adversarial random forests for efficient and robust conditional feature importance estimation with minimal tuning.

Findings

cARFi effectively captures conditional feature importance in various datasets.

The method requires little tuning and adapts to different importance notions.

Statistical tests can be integrated to assess feature importance significance.

Abstract

This paper proposes a method for measuring conditional feature importance via generative modeling. In explainable artificial intelligence (XAI), conditional feature importance assesses the impact of a feature on a prediction model's performance given the information of other features. Model-agnostic post hoc methods to do so typically evaluate changes in the predictive performance under on-manifold feature value manipulations. Such procedures require creating feature values that respect conditional feature distributions, which can be challenging in practice. Recent advancements in generative modeling can facilitate this. For tabular data, which may consist of both categorical and continuous features, the adversarial random forest (ARF) stands out as a generative model that can generate on-manifold data points without requiring intensive tuning efforts or computational resources, making it a promising candidate model for subroutines in XAI methods. This paper proposes cARFi (conditional ARF feature importance), a method for measuring conditional feature importance through feature values sampled from ARF-estimated conditional distributions. cARFi requires only little tuning to yield robust importance scores that can flexibly adapt for conditional or marginal notions of feature importance, including straightforward extensions to condition on feature subsets and allows for inferring the significance of feature importances through statistical tests.