Minimising changes to audit when updating decision trees

TL;DR

This paper introduces a greedy algorithm for updating decision trees that minimizes the number of changes needed for human auditing, balancing accuracy and audit effort.

Contribution

It presents a novel method for updating decision trees with minimal changes, improving interpretability during model updates.

Findings

Balances accuracy and audit effort effectively

Outperforms existing update methods in minimal change criteria

Maintains high interpretability during updates

Abstract

Interpretable models are important, but what happens when the model is updated on new training data? We propose an algorithm for updating a decision tree while minimising the number of changes to the tree that a human would need to audit. We achieve this via a greedy approach that incorporates the number of changes to the tree as part of the objective function. We compare our algorithm to existing methods and show that it sits in a sweet spot between final accuracy and number of changes to audit.