Permutation-Based Causal Structure Learning with Unknown Intervention Targets

Chandler Squires, Yuhao Wang, Caroline Uhler
Proceedings of the 36th Conference on Uncertainty in Artificial Intelligence (UAI), PMLR 124:1039-1048, 2020.

Abstract

We consider the problem of estimating causal DAG models from a mix of observational and interventional data, when the intervention targets are partially or completely unknown. This problem is highly relevant for example in genomics, since gene knockout technologies are known to have off-target effects. We characterize the interventional Markov equivalence class of DAGs that can be identified from interventional data with unknown intervention targets. In addition, we propose a provably consistent algorithm for learning the interventional Markov equivalence class from such data. The proposed algorithm greedily searches over the space of permutations to minimize a novel score function. The algorithm is nonparametric, which is particularly important for applications to genomics, where the relationships between variables are often non-linear and the distribution non-Gaussian. We demonstrate the performance of our algorithm on synthetic and biological datasets. Links to an implementation of our algorithm and to a reproducible code base for our experiments can be found at https://uhlerlab.github.io/causaldag/utigsp.

Cite this Paper

BibTeX
@InProceedings{pmlr-v124-squires20a, title = {Permutation-Based Causal Structure Learning with Unknown Intervention Targets}, author = {Squires, Chandler and Wang, Yuhao and Uhler, Caroline}, booktitle = {Proceedings of the 36th Conference on Uncertainty in Artificial Intelligence (UAI)}, pages = {1039--1048}, year = {2020}, editor = {Peters, Jonas and Sontag, David}, volume = {124}, series = {Proceedings of Machine Learning Research}, month = {03--06 Aug}, publisher = {PMLR}, pdf = {http://proceedings.mlr.press/v124/squires20a/squires20a.pdf}, url = {https://proceedings.mlr.press/v124/squires20a.html}, abstract = {We consider the problem of estimating causal DAG models from a mix of observational and interventional data, when the intervention targets are partially or completely unknown. This problem is highly relevant for example in genomics, since gene knockout technologies are known to have off-target effects. We characterize the interventional Markov equivalence class of DAGs that can be identified from interventional data with unknown intervention targets. In addition, we propose a provably consistent algorithm for learning the interventional Markov equivalence class from such data. The proposed algorithm greedily searches over the space of permutations to minimize a novel score function. The algorithm is nonparametric, which is particularly important for applications to genomics, where the relationships between variables are often non-linear and the distribution non-Gaussian. We demonstrate the performance of our algorithm on synthetic and biological datasets. Links to an implementation of our algorithm and to a reproducible code base for our experiments can be found at https://uhlerlab.github.io/causaldag/utigsp.} }
Endnote
%0 Conference Paper %T Permutation-Based Causal Structure Learning with Unknown Intervention Targets %A Chandler Squires %A Yuhao Wang %A Caroline Uhler %B Proceedings of the 36th Conference on Uncertainty in Artificial Intelligence (UAI) %C Proceedings of Machine Learning Research %D 2020 %E Jonas Peters %E David Sontag %F pmlr-v124-squires20a %I PMLR %P 1039--1048 %U https://proceedings.mlr.press/v124/squires20a.html %V 124 %X We consider the problem of estimating causal DAG models from a mix of observational and interventional data, when the intervention targets are partially or completely unknown. This problem is highly relevant for example in genomics, since gene knockout technologies are known to have off-target effects. We characterize the interventional Markov equivalence class of DAGs that can be identified from interventional data with unknown intervention targets. In addition, we propose a provably consistent algorithm for learning the interventional Markov equivalence class from such data. The proposed algorithm greedily searches over the space of permutations to minimize a novel score function. The algorithm is nonparametric, which is particularly important for applications to genomics, where the relationships between variables are often non-linear and the distribution non-Gaussian. We demonstrate the performance of our algorithm on synthetic and biological datasets. Links to an implementation of our algorithm and to a reproducible code base for our experiments can be found at https://uhlerlab.github.io/causaldag/utigsp.
APA
Squires, C., Wang, Y. & Uhler, C.. (2020). Permutation-Based Causal Structure Learning with Unknown Intervention Targets. Proceedings of the 36th Conference on Uncertainty in Artificial Intelligence (UAI), in Proceedings of Machine Learning Research 124:1039-1048 Available from https://proceedings.mlr.press/v124/squires20a.html.

Related Material