Computing Abductive Explanations for Boosted Trees

Gilles Audemard, Jean-Marie Lagniez, Pierre Marquis, Nicolas Szczepanski
Proceedings of The 26th International Conference on Artificial Intelligence and Statistics, PMLR 206:4699-4711, 2023.

Abstract

Boosted trees is a dominant ML model, exhibiting high accuracy. However, boosted trees are hardly intelligible, and this is a problem whenever they are used in safety-critical applications. Indeed, in such a context, provably sound explanations for the predictions made are expected. Recent work have shown how subset-minimal abductive explanations can be derived for boosted trees, using automated reasoning techniques. However, the generation of such well-founded explanations is intractable in the general case. To improve the scalability of their generation, we introduce the notion of tree-specific explanation for a boosted tree. We show that tree-specific explanations are provably sound abductive explanations that can be computed in polynomial time. We also explain how to derive a subset-minimal abductive explanation from a tree-specific explanation. Experiments on various datasets show the computational benefits of leveraging tree-specific explanations for deriving subset-minimal abductive explanations.

Cite this Paper


BibTeX
@InProceedings{pmlr-v206-audemard23a, title = {Computing Abductive Explanations for Boosted Trees}, author = {Audemard, Gilles and Lagniez, Jean-Marie and Marquis, Pierre and Szczepanski, Nicolas}, booktitle = {Proceedings of The 26th International Conference on Artificial Intelligence and Statistics}, pages = {4699--4711}, year = {2023}, editor = {Ruiz, Francisco and Dy, Jennifer and van de Meent, Jan-Willem}, volume = {206}, series = {Proceedings of Machine Learning Research}, month = {25--27 Apr}, publisher = {PMLR}, pdf = {https://proceedings.mlr.press/v206/audemard23a/audemard23a.pdf}, url = {https://proceedings.mlr.press/v206/audemard23a.html}, abstract = {Boosted trees is a dominant ML model, exhibiting high accuracy. However, boosted trees are hardly intelligible, and this is a problem whenever they are used in safety-critical applications. Indeed, in such a context, provably sound explanations for the predictions made are expected. Recent work have shown how subset-minimal abductive explanations can be derived for boosted trees, using automated reasoning techniques. However, the generation of such well-founded explanations is intractable in the general case. To improve the scalability of their generation, we introduce the notion of tree-specific explanation for a boosted tree. We show that tree-specific explanations are provably sound abductive explanations that can be computed in polynomial time. We also explain how to derive a subset-minimal abductive explanation from a tree-specific explanation. Experiments on various datasets show the computational benefits of leveraging tree-specific explanations for deriving subset-minimal abductive explanations.} }
Endnote
%0 Conference Paper %T Computing Abductive Explanations for Boosted Trees %A Gilles Audemard %A Jean-Marie Lagniez %A Pierre Marquis %A Nicolas Szczepanski %B Proceedings of The 26th International Conference on Artificial Intelligence and Statistics %C Proceedings of Machine Learning Research %D 2023 %E Francisco Ruiz %E Jennifer Dy %E Jan-Willem van de Meent %F pmlr-v206-audemard23a %I PMLR %P 4699--4711 %U https://proceedings.mlr.press/v206/audemard23a.html %V 206 %X Boosted trees is a dominant ML model, exhibiting high accuracy. However, boosted trees are hardly intelligible, and this is a problem whenever they are used in safety-critical applications. Indeed, in such a context, provably sound explanations for the predictions made are expected. Recent work have shown how subset-minimal abductive explanations can be derived for boosted trees, using automated reasoning techniques. However, the generation of such well-founded explanations is intractable in the general case. To improve the scalability of their generation, we introduce the notion of tree-specific explanation for a boosted tree. We show that tree-specific explanations are provably sound abductive explanations that can be computed in polynomial time. We also explain how to derive a subset-minimal abductive explanation from a tree-specific explanation. Experiments on various datasets show the computational benefits of leveraging tree-specific explanations for deriving subset-minimal abductive explanations.
APA
Audemard, G., Lagniez, J., Marquis, P. & Szczepanski, N.. (2023). Computing Abductive Explanations for Boosted Trees. Proceedings of The 26th International Conference on Artificial Intelligence and Statistics, in Proceedings of Machine Learning Research 206:4699-4711 Available from https://proceedings.mlr.press/v206/audemard23a.html.

Related Material