Variance Minimization in the Wasserstein Space for Invariant Causal Prediction

Guillaume G. Martinet, Alexander Strzalkowski, Barbara Engelhardt
Proceedings of The 25th International Conference on Artificial Intelligence and Statistics, PMLR 151:8803-8851, 2022.

Abstract

Selecting powerful predictors for an outcome is a cornerstone task for machine learning. However, some types of questions can only be answered by identifying the predictors that causally affect the outcome. A recent approach to this causal inference problem leverages the invariance property of a causal mechanism across differing experimental environments (Peters et al., 2016; Heinze-Deml et al., 2018). This method, invariant causal prediction (ICP), has a substantial computational defect – the runtime scales exponentially with the number of possible causal variables. In this work, we show that the approach taken in ICP may be reformulated as a series of nonparametric tests that scales linearly in the number of predictors. Each of these tests relies on the minimization of a novel loss function – the Wasserstein variance – that is derived from tools in optimal transport theory and is used to quantify distributional variability across environments. We prove under mild assumptions that our method is able to recover the set of identifiable direct causes, and we demonstrate in our experiments that it is competitive with other benchmark causal discovery algorithms.

Cite this Paper

BibTeX
@InProceedings{pmlr-v151-martinet22a, title = { Variance Minimization in the Wasserstein Space for Invariant Causal Prediction }, author = {Martinet, Guillaume G. and Strzalkowski, Alexander and Engelhardt, Barbara}, booktitle = {Proceedings of The 25th International Conference on Artificial Intelligence and Statistics}, pages = {8803--8851}, year = {2022}, editor = {Camps-Valls, Gustau and Ruiz, Francisco J. R. and Valera, Isabel}, volume = {151}, series = {Proceedings of Machine Learning Research}, month = {28--30 Mar}, publisher = {PMLR}, pdf = {https://proceedings.mlr.press/v151/martinet22a/martinet22a.pdf}, url = {https://proceedings.mlr.press/v151/martinet22a.html}, abstract = { Selecting powerful predictors for an outcome is a cornerstone task for machine learning. However, some types of questions can only be answered by identifying the predictors that causally affect the outcome. A recent approach to this causal inference problem leverages the invariance property of a causal mechanism across differing experimental environments (Peters et al., 2016; Heinze-Deml et al., 2018). This method, invariant causal prediction (ICP), has a substantial computational defect – the runtime scales exponentially with the number of possible causal variables. In this work, we show that the approach taken in ICP may be reformulated as a series of nonparametric tests that scales linearly in the number of predictors. Each of these tests relies on the minimization of a novel loss function – the Wasserstein variance – that is derived from tools in optimal transport theory and is used to quantify distributional variability across environments. We prove under mild assumptions that our method is able to recover the set of identifiable direct causes, and we demonstrate in our experiments that it is competitive with other benchmark causal discovery algorithms. } }
Endnote
%0 Conference Paper %T Variance Minimization in the Wasserstein Space for Invariant Causal Prediction %A Guillaume G. Martinet %A Alexander Strzalkowski %A Barbara Engelhardt %B Proceedings of The 25th International Conference on Artificial Intelligence and Statistics %C Proceedings of Machine Learning Research %D 2022 %E Gustau Camps-Valls %E Francisco J. R. Ruiz %E Isabel Valera %F pmlr-v151-martinet22a %I PMLR %P 8803--8851 %U https://proceedings.mlr.press/v151/martinet22a.html %V 151 %X Selecting powerful predictors for an outcome is a cornerstone task for machine learning. However, some types of questions can only be answered by identifying the predictors that causally affect the outcome. A recent approach to this causal inference problem leverages the invariance property of a causal mechanism across differing experimental environments (Peters et al., 2016; Heinze-Deml et al., 2018). This method, invariant causal prediction (ICP), has a substantial computational defect – the runtime scales exponentially with the number of possible causal variables. In this work, we show that the approach taken in ICP may be reformulated as a series of nonparametric tests that scales linearly in the number of predictors. Each of these tests relies on the minimization of a novel loss function – the Wasserstein variance – that is derived from tools in optimal transport theory and is used to quantify distributional variability across environments. We prove under mild assumptions that our method is able to recover the set of identifiable direct causes, and we demonstrate in our experiments that it is competitive with other benchmark causal discovery algorithms.
APA
Martinet, G.G., Strzalkowski, A. & Engelhardt, B.. (2022). Variance Minimization in the Wasserstein Space for Invariant Causal Prediction . Proceedings of The 25th International Conference on Artificial Intelligence and Statistics, in Proceedings of Machine Learning Research 151:8803-8851 Available from https://proceedings.mlr.press/v151/martinet22a.html.

Related Material