Saved in:
Bibliographic Details
Main Authors: Simmons, Anj, Barnett, Scott, Chaudhuri, Anupam, Singh, Sankhya, Sivasothy, Shangeetha
Format: Preprint
Published: 2024
Subjects:
Online Access:https://arxiv.org/abs/2408.16321
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866917762218065920
author Simmons, Anj
Barnett, Scott
Chaudhuri, Anupam
Singh, Sankhya
Sivasothy, Shangeetha
author_facet Simmons, Anj
Barnett, Scott
Chaudhuri, Anupam
Singh, Sankhya
Sivasothy, Shangeetha
contents 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.
format Preprint
id arxiv_https___arxiv_org_abs_2408_16321
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Minimising changes to audit when updating decision trees
Simmons, Anj
Barnett, Scott
Chaudhuri, Anupam
Singh, Sankhya
Sivasothy, Shangeetha
Machine Learning
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.
title Minimising changes to audit when updating decision trees
topic Machine Learning
url https://arxiv.org/abs/2408.16321