Estimation of Counterfactual Interventions under Uncertainties

Juliane Weilbach, Sebastian Gerwinn, Melih Kandemir, Martin Fraenzle
Proceedings of the 15th Asian Conference on Machine Learning, PMLR 222:1463-1478, 2024.

Abstract

Counterfactual analysis is intuitively performed by humans on a daily basis eg. ”What should I have done differently to get the loan approved?”. Such counterfactual questions also steer the formulation of scientific hypotheses. More formally it provides insights about potential improvements of a system by inferring the effects of hypothetical interventions into a past observation of the system’s behaviour which plays a prominent role in a variety of industrial applications. Due to the hypothetical nature of such analysis, counterfactual distributions are inherently ambiguous. This ambiguity is particularly challenging in continuous settings in which a continuum of explanations exist for the same observation. In this paper, we address this problem by following a hierarchical Bayesian approach which explicitly models such uncertainty. In particular, we derive counterfactual distributions for a Bayesian Warped Gaussian Process thereby allowing for non-Gaussian distributions and non-additive noise. We illustrate the properties of our approach on a synthetic and on a semi-synthetic example and show its performance when used within an algorithmic recourse downstream task.

Cite this Paper

BibTeX
@InProceedings{pmlr-v222-weilbach24a, title = {Estimation of Counterfactual Interventions under Uncertainties}, author = {Weilbach, Juliane and Gerwinn, Sebastian and Kandemir, Melih and Fraenzle, Martin}, booktitle = {Proceedings of the 15th Asian Conference on Machine Learning}, pages = {1463--1478}, year = {2024}, editor = {Yanıkoğlu, Berrin and Buntine, Wray}, volume = {222}, series = {Proceedings of Machine Learning Research}, month = {11--14 Nov}, publisher = {PMLR}, pdf = {https://proceedings.mlr.press/v222/weilbach24a/weilbach24a.pdf}, url = {https://proceedings.mlr.press/v222/weilbach24a.html}, abstract = {Counterfactual analysis is intuitively performed by humans on a daily basis eg. ”What should I have done differently to get the loan approved?”. Such counterfactual questions also steer the formulation of scientific hypotheses. More formally it provides insights about potential improvements of a system by inferring the effects of hypothetical interventions into a past observation of the system’s behaviour which plays a prominent role in a variety of industrial applications. Due to the hypothetical nature of such analysis, counterfactual distributions are inherently ambiguous. This ambiguity is particularly challenging in continuous settings in which a continuum of explanations exist for the same observation. In this paper, we address this problem by following a hierarchical Bayesian approach which explicitly models such uncertainty. In particular, we derive counterfactual distributions for a Bayesian Warped Gaussian Process thereby allowing for non-Gaussian distributions and non-additive noise. We illustrate the properties of our approach on a synthetic and on a semi-synthetic example and show its performance when used within an algorithmic recourse downstream task.} }
Endnote
%0 Conference Paper %T Estimation of Counterfactual Interventions under Uncertainties %A Juliane Weilbach %A Sebastian Gerwinn %A Melih Kandemir %A Martin Fraenzle %B Proceedings of the 15th Asian Conference on Machine Learning %C Proceedings of Machine Learning Research %D 2024 %E Berrin Yanıkoğlu %E Wray Buntine %F pmlr-v222-weilbach24a %I PMLR %P 1463--1478 %U https://proceedings.mlr.press/v222/weilbach24a.html %V 222 %X Counterfactual analysis is intuitively performed by humans on a daily basis eg. ”What should I have done differently to get the loan approved?”. Such counterfactual questions also steer the formulation of scientific hypotheses. More formally it provides insights about potential improvements of a system by inferring the effects of hypothetical interventions into a past observation of the system’s behaviour which plays a prominent role in a variety of industrial applications. Due to the hypothetical nature of such analysis, counterfactual distributions are inherently ambiguous. This ambiguity is particularly challenging in continuous settings in which a continuum of explanations exist for the same observation. In this paper, we address this problem by following a hierarchical Bayesian approach which explicitly models such uncertainty. In particular, we derive counterfactual distributions for a Bayesian Warped Gaussian Process thereby allowing for non-Gaussian distributions and non-additive noise. We illustrate the properties of our approach on a synthetic and on a semi-synthetic example and show its performance when used within an algorithmic recourse downstream task.
APA
Weilbach, J., Gerwinn, S., Kandemir, M. & Fraenzle, M.. (2024). Estimation of Counterfactual Interventions under Uncertainties. Proceedings of the 15th Asian Conference on Machine Learning, in Proceedings of Machine Learning Research 222:1463-1478 Available from https://proceedings.mlr.press/v222/weilbach24a.html.

Related Material