Multi-Granular Evaluation of Diverse Counterfactual Explanations

Research output: Chapter in Book/Report/Conference proceedingConference Contribution (Conference Proceeding)

21 Downloads (Pure)


As a popular approach in Explainable AI (XAI), an increasing number of counterfactual explanation algorithms have been proposed in the context of making machine learning classifiers more trustworthy and transparent. This paper reports our evaluations of algorithms that can output diverse counterfactuals for one instance. We first evaluate the performance of DiCE-Random, DiCE-KDTree, DiCE-Genetic and Alibi-CFRL, taking XGBoost as the machine learning model for binary classification problems. Then, we compare their suggested feature changes with feature importance by SHAP. Moreover, our study highlights that synthetic counterfactuals, drawn from the input domain but not necessarily the training data, outperform native counterfactuals from the training data regarding data privacy and validity. This research aims to guide practitioners in choosing the most suitable algorithm for generating diverse counterfactual explanations.
Original languageEnglish
Title of host publicationProceedings of the 16th International Conference on Agents and Artificial Intelligence
EditorsAna Paula Rocha, Luc Steels, Jaap van den Herik
Number of pages12
ISBN (Electronic)978-989-758-680-4
Publication statusPublished - 3 Mar 2024
EventICAART2024 : 16th International Conference on Agents and Artificial Intelligence - Italy, Rome, Italy
Duration: 24 Feb 202426 Feb 2024
Conference number: 16

Publication series

NameInternational Conference on Agents and Artificial Intelligence
ISSN (Print)2184-3589
ISSN (Electronic)2184-433X


Abbreviated titleICAART2024
Internet address

Bibliographical note

Publisher Copyright:
© 2024 by SCITEPRESS – Science and Technology Publications, Lda.


  • Counterfactual Explanations
  • Explainable AI


Dive into the research topics of 'Multi-Granular Evaluation of Diverse Counterfactual Explanations'. Together they form a unique fingerprint.

Cite this