Example-based Explanations for Random Forests using Machine Unlearning

TL;DR

This paper introduces FairDebugger, a system that uses machine unlearning and data mining techniques to identify training data responsible for fairness violations in random forest classifiers.

Contribution

It presents a novel method combining machine unlearning and frequent itemset mining to explain and debug fairness issues in tree-based models.

Findings

FairDebugger effectively identifies data subsets linked to unfair outcomes.

The explanations align with prior domain knowledge and insights.

The approach is validated on three real-world datasets.

Abstract

Tree-based machine learning models, such as decision trees and random forests, have been hugely successful in classification tasks primarily because of their predictive power in supervised learning tasks and ease of interpretation. Despite their popularity and power, these models have been found to produce unexpected or discriminatory outcomes. Given their overwhelming success for most tasks, it is of interest to identify sources of their unexpected and discriminatory behavior. However, there has not been much work on understanding and debugging tree-based classifiers in the context of fairness. We introduce FairDebugger, a system that utilizes recent advances in machine unlearning research to identify training data subsets responsible for instances of fairness violations in the outcomes of a random forest classifier. FairDebugger generates top-$k$ explanations (in the form of coherent training data subsets) for model unfairness. Toward this goal, FairDebugger first utilizes machine unlearning to estimate the change in the tree structures of the random forest when parts of the underlying training data are removed, and then leverages the Apriori algorithm from frequent itemset mining to reduce the subset search space. We empirically evaluate our approach on three real-world datasets, and demonstrate that the explanations generated by FairDebugger are consistent with insights from prior studies on these datasets.