On Data Manifolds Entailed by Structural Causal Models

Ricardo Dominguez-Olmedo, Amir-Hossein Karimi, Georgios Arvanitidis, Bernhard Schölkopf
Proceedings of the 40th International Conference on Machine Learning, PMLR 202:8188-8201, 2023.

Abstract

The geometric structure of data is an important inductive bias in machine learning. In this work, we characterize the data manifolds entailed by structural causal models. The strengths of the proposed framework are twofold: firstly, the geometric structure of the data manifolds is causally informed, and secondly, it enables causal reasoning about the data manifolds in an interventional and a counterfactual sense. We showcase the versatility of the proposed framework by applying it to the generation of causally-grounded counterfactual explanations for machine learning classifiers, measuring distances along the data manifold in a differential geometric-principled manner.

Cite this Paper

BibTeX
@InProceedings{pmlr-v202-dominguez-olmedo23a, title = {On Data Manifolds Entailed by Structural Causal Models}, author = {Dominguez-Olmedo, Ricardo and Karimi, Amir-Hossein and Arvanitidis, Georgios and Sch\"{o}lkopf, Bernhard}, booktitle = {Proceedings of the 40th International Conference on Machine Learning}, pages = {8188--8201}, year = {2023}, editor = {Krause, Andreas and Brunskill, Emma and Cho, Kyunghyun and Engelhardt, Barbara and Sabato, Sivan and Scarlett, Jonathan}, volume = {202}, series = {Proceedings of Machine Learning Research}, month = {23--29 Jul}, publisher = {PMLR}, pdf = {https://proceedings.mlr.press/v202/dominguez-olmedo23a/dominguez-olmedo23a.pdf}, url = {https://proceedings.mlr.press/v202/dominguez-olmedo23a.html}, abstract = {The geometric structure of data is an important inductive bias in machine learning. In this work, we characterize the data manifolds entailed by structural causal models. The strengths of the proposed framework are twofold: firstly, the geometric structure of the data manifolds is causally informed, and secondly, it enables causal reasoning about the data manifolds in an interventional and a counterfactual sense. We showcase the versatility of the proposed framework by applying it to the generation of causally-grounded counterfactual explanations for machine learning classifiers, measuring distances along the data manifold in a differential geometric-principled manner.} }
Endnote
%0 Conference Paper %T On Data Manifolds Entailed by Structural Causal Models %A Ricardo Dominguez-Olmedo %A Amir-Hossein Karimi %A Georgios Arvanitidis %A Bernhard Schölkopf %B Proceedings of the 40th International Conference on Machine Learning %C Proceedings of Machine Learning Research %D 2023 %E Andreas Krause %E Emma Brunskill %E Kyunghyun Cho %E Barbara Engelhardt %E Sivan Sabato %E Jonathan Scarlett %F pmlr-v202-dominguez-olmedo23a %I PMLR %P 8188--8201 %U https://proceedings.mlr.press/v202/dominguez-olmedo23a.html %V 202 %X The geometric structure of data is an important inductive bias in machine learning. In this work, we characterize the data manifolds entailed by structural causal models. The strengths of the proposed framework are twofold: firstly, the geometric structure of the data manifolds is causally informed, and secondly, it enables causal reasoning about the data manifolds in an interventional and a counterfactual sense. We showcase the versatility of the proposed framework by applying it to the generation of causally-grounded counterfactual explanations for machine learning classifiers, measuring distances along the data manifold in a differential geometric-principled manner.
APA
Dominguez-Olmedo, R., Karimi, A., Arvanitidis, G. & Schölkopf, B.. (2023). On Data Manifolds Entailed by Structural Causal Models. Proceedings of the 40th International Conference on Machine Learning, in Proceedings of Machine Learning Research 202:8188-8201 Available from https://proceedings.mlr.press/v202/dominguez-olmedo23a.html.

Related Material