Covered Information Disentanglement: Model Transparency via Unbiased Permutation Importance

TL;DR

This paper introduces Covered Information Disentanglement (CID), a novel method that corrects permutation importance by accounting for feature overlap, enhancing model transparency especially in sensitive domains like medicine.

Contribution

The paper proposes CID, a new approach that adjusts permutation importance for feature overlap, improving interpretability of machine learning models in complex data scenarios.

Findings

CID effectively corrects importance scores in toy datasets.

CID improves feature importance estimation in medical data.

Efficient computation of CID with Markov random fields.

Abstract

Model transparency is a prerequisite in many domains and an increasingly popular area in machine learning research. In the medical domain, for instance, unveiling the mechanisms behind a disease often has higher priority than the diagnostic itself since it might dictate or guide potential treatments and research directions. One of the most popular approaches to explain model global predictions is the permutation importance where the performance on permuted data is benchmarked against the baseline. However, this method and other related approaches will undervalue the importance of a feature in the presence of covariates since these cover part of its provided information. To address this issue, we propose Covered Information Disentanglement (CID), a method that considers all feature information overlap to correct the values provided by permutation importance. We further show how to compute CID efficiently when coupled with Markov random fields. We demonstrate its efficacy in adjusting permutation importance first on a controlled toy dataset and discuss its effect on real-world medical data.