Counterfactual Metarules for Local and Global Recourse

Tom Bewley, Salim I. Amoukou, Saumitra Mishra, Daniele Magazzeni, Manuela Veloso
Proceedings of the 41st International Conference on Machine Learning, PMLR 235:3707-3724, 2024.

Abstract

We introduce T-CREx, a novel model-agnostic method for local and global counterfactual explanation (CE), which summarises recourse options for both individuals and groups in the form of generalised rules. It leverages tree-based surrogate models to learn the counterfactual rules, alongside metarules denoting their regimes of optimality, providing both a global analysis of model behaviour and diverse recourse options for users. Experiments indicate that T-CREx achieves superior aggregate performance over existing rule-based baselines on a range of CE desiderata, while being orders of magnitude faster to run.

Cite this Paper

BibTeX
@InProceedings{pmlr-v235-bewley24a, title = {Counterfactual Metarules for Local and Global Recourse}, author = {Bewley, Tom and I. Amoukou, Salim and Mishra, Saumitra and Magazzeni, Daniele and Veloso, Manuela}, booktitle = {Proceedings of the 41st International Conference on Machine Learning}, pages = {3707--3724}, year = {2024}, editor = {Salakhutdinov, Ruslan and Kolter, Zico and Heller, Katherine and Weller, Adrian and Oliver, Nuria and Scarlett, Jonathan and Berkenkamp, Felix}, volume = {235}, series = {Proceedings of Machine Learning Research}, month = {21--27 Jul}, publisher = {PMLR}, pdf = {https://raw.githubusercontent.com/mlresearch/v235/main/assets/bewley24a/bewley24a.pdf}, url = {https://proceedings.mlr.press/v235/bewley24a.html}, abstract = {We introduce T-CREx, a novel model-agnostic method for local and global counterfactual explanation (CE), which summarises recourse options for both individuals and groups in the form of generalised rules. It leverages tree-based surrogate models to learn the counterfactual rules, alongside metarules denoting their regimes of optimality, providing both a global analysis of model behaviour and diverse recourse options for users. Experiments indicate that T-CREx achieves superior aggregate performance over existing rule-based baselines on a range of CE desiderata, while being orders of magnitude faster to run.} }
Endnote
%0 Conference Paper %T Counterfactual Metarules for Local and Global Recourse %A Tom Bewley %A Salim I. Amoukou %A Saumitra Mishra %A Daniele Magazzeni %A Manuela Veloso %B Proceedings of the 41st International Conference on Machine Learning %C Proceedings of Machine Learning Research %D 2024 %E Ruslan Salakhutdinov %E Zico Kolter %E Katherine Heller %E Adrian Weller %E Nuria Oliver %E Jonathan Scarlett %E Felix Berkenkamp %F pmlr-v235-bewley24a %I PMLR %P 3707--3724 %U https://proceedings.mlr.press/v235/bewley24a.html %V 235 %X We introduce T-CREx, a novel model-agnostic method for local and global counterfactual explanation (CE), which summarises recourse options for both individuals and groups in the form of generalised rules. It leverages tree-based surrogate models to learn the counterfactual rules, alongside metarules denoting their regimes of optimality, providing both a global analysis of model behaviour and diverse recourse options for users. Experiments indicate that T-CREx achieves superior aggregate performance over existing rule-based baselines on a range of CE desiderata, while being orders of magnitude faster to run.
APA
Bewley, T., I. Amoukou, S., Mishra, S., Magazzeni, D. & Veloso, M.. (2024). Counterfactual Metarules for Local and Global Recourse. Proceedings of the 41st International Conference on Machine Learning, in Proceedings of Machine Learning Research 235:3707-3724 Available from https://proceedings.mlr.press/v235/bewley24a.html.

Related Material