A Differential-Evolution-Based Approach to Extract Univariate Decision Trees From Black-Box Models Using Tabular Data

The growing demand for complex machine learning models has increased the use of black-box models, such as random forests and artificial neural networks, posing significant challenges regarding explainability and interpretability. This manuscript addresses the critical problem of understanding and in...

Full description

Saved in:
Bibliographic Details
Main Authors: Rafael Rivera-Lopez, Hector G. Ceballos
Format: Article
Language:English
Published: IEEE 2024-01-01
Series:IEEE Access
Subjects:
Online Access:https://ieeexplore.ieee.org/document/10753588/
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1850059744538001408
author Rafael Rivera-Lopez
Hector G. Ceballos
author_facet Rafael Rivera-Lopez
Hector G. Ceballos
author_sort Rafael Rivera-Lopez
collection DOAJ
description The growing demand for complex machine learning models has increased the use of black-box models, such as random forests and artificial neural networks, posing significant challenges regarding explainability and interpretability. This manuscript addresses the critical problem of understanding and interpreting decisions from these opaque models, as a lack of interpretability can hinder their adoption in sensitive applications. To tackle this issue, we propose an evolutionary approach to induce univariate decision trees that accurately mimic the behavior of black-box models using tabular data. Our method employs two differential evolution algorithm variants, focusing on building univariate decision trees to enhance model explainability. Key contributions of this work include the development of a fitness function that balances accuracy with tree compactness to reduce overfitting and improve explanability. Additionally, we introduce a novel selection scheme that evaluates candidate solutions using synthetic instances, further enhancing the robustness against variance of the decision trees. Experimental results demonstrate that the proposed approach yields more precise and compact decision trees than traditional methods, significantly improving the explainability of complex machine learning models.
format Article
id doaj-art-e144ce46734a4c70b948e857ecf8bb30
institution DOAJ
issn 2169-3536
language English
publishDate 2024-01-01
publisher IEEE
record_format Article
series IEEE Access
spelling doaj-art-e144ce46734a4c70b948e857ecf8bb302025-08-20T02:50:48ZengIEEEIEEE Access2169-35362024-01-011216985016986810.1109/ACCESS.2024.349890710753588A Differential-Evolution-Based Approach to Extract Univariate Decision Trees From Black-Box Models Using Tabular DataRafael Rivera-Lopez0https://orcid.org/0000-0002-5254-4195Hector G. Ceballos1https://orcid.org/0000-0002-2460-3442Departamento de Sistemas y Computación, Tecnológico Nacional de México, Instituto Tecnológico de Veracruz, Veracruz, MexicoInstitute for the Future of Education, Tecnológico de Monterrey, Monterrey, MexicoThe growing demand for complex machine learning models has increased the use of black-box models, such as random forests and artificial neural networks, posing significant challenges regarding explainability and interpretability. This manuscript addresses the critical problem of understanding and interpreting decisions from these opaque models, as a lack of interpretability can hinder their adoption in sensitive applications. To tackle this issue, we propose an evolutionary approach to induce univariate decision trees that accurately mimic the behavior of black-box models using tabular data. Our method employs two differential evolution algorithm variants, focusing on building univariate decision trees to enhance model explainability. Key contributions of this work include the development of a fitness function that balances accuracy with tree compactness to reduce overfitting and improve explanability. Additionally, we introduce a novel selection scheme that evaluates candidate solutions using synthetic instances, further enhancing the robustness against variance of the decision trees. Experimental results demonstrate that the proposed approach yields more precise and compact decision trees than traditional methods, significantly improving the explainability of complex machine learning models.https://ieeexplore.ieee.org/document/10753588/Agnostic modelexplainable artificial intelligenceevolutionary computationdecision trees
spellingShingle Rafael Rivera-Lopez
Hector G. Ceballos
A Differential-Evolution-Based Approach to Extract Univariate Decision Trees From Black-Box Models Using Tabular Data
IEEE Access
Agnostic model
explainable artificial intelligence
evolutionary computation
decision trees
title A Differential-Evolution-Based Approach to Extract Univariate Decision Trees From Black-Box Models Using Tabular Data
title_full A Differential-Evolution-Based Approach to Extract Univariate Decision Trees From Black-Box Models Using Tabular Data
title_fullStr A Differential-Evolution-Based Approach to Extract Univariate Decision Trees From Black-Box Models Using Tabular Data
title_full_unstemmed A Differential-Evolution-Based Approach to Extract Univariate Decision Trees From Black-Box Models Using Tabular Data
title_short A Differential-Evolution-Based Approach to Extract Univariate Decision Trees From Black-Box Models Using Tabular Data
title_sort differential evolution based approach to extract univariate decision trees from black box models using tabular data
topic Agnostic model
explainable artificial intelligence
evolutionary computation
decision trees
url https://ieeexplore.ieee.org/document/10753588/
work_keys_str_mv AT rafaelriveralopez adifferentialevolutionbasedapproachtoextractunivariatedecisiontreesfromblackboxmodelsusingtabulardata
AT hectorgceballos adifferentialevolutionbasedapproachtoextractunivariatedecisiontreesfromblackboxmodelsusingtabulardata
AT rafaelriveralopez differentialevolutionbasedapproachtoextractunivariatedecisiontreesfromblackboxmodelsusingtabulardata
AT hectorgceballos differentialevolutionbasedapproachtoextractunivariatedecisiontreesfromblackboxmodelsusingtabulardata