Improving the Validity of Decision Trees as Explanations

Fuente: arXiv
Salvato in:
Dettagli Bibliografici
Autori principali: Nemecek, Jiri, Pevny, Tomas, Marecek, Jakub
Natura: Preprint
Pubblicazione: 2023
Soggetti:
Accesso online:
Tags: Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
_version_ 1866909216176865280
author Nemecek, Jiri
Pevny, Tomas
Marecek, Jakub
author_facet Nemecek, Jiri
Pevny, Tomas
Marecek, Jakub
contents In classification and forecasting with tabular data, one often utilizes tree-based models. Those can be competitive with deep neural networks on tabular data and, under some conditions, explainable. The explainability depends on the depth of the tree and the accuracy in each leaf of the tree. We point out that decision trees containing leaves with unbalanced accuracy can provide misleading explanations. Low-accuracy leaves give less valid explanations, which could be interpreted as unfairness among subgroups utilizing these explanations. Here, we train a shallow tree with the objective of minimizing the maximum misclassification error across all leaf nodes. The shallow tree provides a global explanation, while the overall statistical performance of the shallow tree can become comparable to state-of-the-art methods (e.g., well-tuned XGBoost) by extending the leaves with further models.
format Preprint
id arxiv_https___arxiv_org_abs_2306_06777
institution arXiv
publishDate 2023
record_format arxiv
spellingShingle Improving the Validity of Decision Trees as Explanations
Nemecek, Jiri
Pevny, Tomas
Marecek, Jakub
Machine Learning
Artificial Intelligence
Optimization and Control
In classification and forecasting with tabular data, one often utilizes tree-based models. Those can be competitive with deep neural networks on tabular data and, under some conditions, explainable. The explainability depends on the depth of the tree and the accuracy in each leaf of the tree. We point out that decision trees containing leaves with unbalanced accuracy can provide misleading explanations. Low-accuracy leaves give less valid explanations, which could be interpreted as unfairness among subgroups utilizing these explanations. Here, we train a shallow tree with the objective of minimizing the maximum misclassification error across all leaf nodes. The shallow tree provides a global explanation, while the overall statistical performance of the shallow tree can become comparable to state-of-the-art methods (e.g., well-tuned XGBoost) by extending the leaves with further models.
title Improving the Validity of Decision Trees as Explanations
topic Machine Learning
Artificial Intelligence
Optimization and Control
url https://arxiv.org/abs/2306.06777