Multi-objective Counterfactuals in Bayesian Classifiers with Estimation of Distribution Algorithms

Daniel Zaragoza-Pellicer, Concha Bielza, Pedro Larrañaga
Proceedings of The 12th International Conference on Probabilistic Graphical Models, PMLR 246:415-426, 2024.

Abstract

Counterfactual explanations are a very popular and effective method to convey interpretability in supervised classification models. These explanations answer the question of which change is needed in the input data to obtain a desired output. Computing good counterfactuals involves achieving some key objectives, such as validity, minimality, similarity or plausibility. Our proposal consists of using estimation of distribution algorithms for approximating counterfactual explanations within Bayesian classifiers. They are experimentally compared with a genetic algorithm, both with a single-objective and with a multi-objective formulation. Different types of Bayesian classifiers will be evaluated to find the differences in their explanations and we will use their results together to provide more accurate explanations. The experiments show how estimation of distribution algorithms are faster and achieve better results with a single-objective whereas they are competitive in the multi-objective version.

Cite this Paper

BibTeX
@InProceedings{pmlr-v246-zaragoza-pellicer24a, title = {Multi-objective Counterfactuals in Bayesian Classifiers with Estimation of Distribution Algorithms}, author = {Zaragoza-Pellicer, Daniel and Bielza, Concha and Larra\~{n}aga, Pedro}, booktitle = {Proceedings of The 12th International Conference on Probabilistic Graphical Models}, pages = {415--426}, year = {2024}, editor = {Kwisthout, Johan and Renooij, Silja}, volume = {246}, series = {Proceedings of Machine Learning Research}, month = {11--13 Sep}, publisher = {PMLR}, pdf = {https://raw.githubusercontent.com/mlresearch/v246/main/assets/zaragoza-pellicer24a/zaragoza-pellicer24a.pdf}, url = {https://proceedings.mlr.press/v246/zaragoza-pellicer24a.html}, abstract = {Counterfactual explanations are a very popular and effective method to convey interpretability in supervised classification models. These explanations answer the question of which change is needed in the input data to obtain a desired output. Computing good counterfactuals involves achieving some key objectives, such as validity, minimality, similarity or plausibility. Our proposal consists of using estimation of distribution algorithms for approximating counterfactual explanations within Bayesian classifiers. They are experimentally compared with a genetic algorithm, both with a single-objective and with a multi-objective formulation. Different types of Bayesian classifiers will be evaluated to find the differences in their explanations and we will use their results together to provide more accurate explanations. The experiments show how estimation of distribution algorithms are faster and achieve better results with a single-objective whereas they are competitive in the multi-objective version.} }
Endnote
%0 Conference Paper %T Multi-objective Counterfactuals in Bayesian Classifiers with Estimation of Distribution Algorithms %A Daniel Zaragoza-Pellicer %A Concha Bielza %A Pedro Larrañaga %B Proceedings of The 12th International Conference on Probabilistic Graphical Models %C Proceedings of Machine Learning Research %D 2024 %E Johan Kwisthout %E Silja Renooij %F pmlr-v246-zaragoza-pellicer24a %I PMLR %P 415--426 %U https://proceedings.mlr.press/v246/zaragoza-pellicer24a.html %V 246 %X Counterfactual explanations are a very popular and effective method to convey interpretability in supervised classification models. These explanations answer the question of which change is needed in the input data to obtain a desired output. Computing good counterfactuals involves achieving some key objectives, such as validity, minimality, similarity or plausibility. Our proposal consists of using estimation of distribution algorithms for approximating counterfactual explanations within Bayesian classifiers. They are experimentally compared with a genetic algorithm, both with a single-objective and with a multi-objective formulation. Different types of Bayesian classifiers will be evaluated to find the differences in their explanations and we will use their results together to provide more accurate explanations. The experiments show how estimation of distribution algorithms are faster and achieve better results with a single-objective whereas they are competitive in the multi-objective version.
APA
Zaragoza-Pellicer, D., Bielza, C. & Larrañaga, P.. (2024). Multi-objective Counterfactuals in Bayesian Classifiers with Estimation of Distribution Algorithms. Proceedings of The 12th International Conference on Probabilistic Graphical Models, in Proceedings of Machine Learning Research 246:415-426 Available from https://proceedings.mlr.press/v246/zaragoza-pellicer24a.html.

Related Material