CREDICI: A Java Library for Causal Inference by Credal Networks

Rafael Cabañas, Alessandro Antonucci, David Huber, Marco Zaffalon
Proceedings of the 10th International Conference on Probabilistic Graphical Models, PMLR 138:597-600, 2020.

Abstract

We present CREDICI, a Java open-source tool for causal inference based on credal networks. Credal networks are an extension of Bayesian networks where local probability mass functions are only constrained to belong to given, so-called credal, sets. CREDICI is based on the recent work of Zaffalon et al. (2020), where an equivalence between Pearl’s structural causal models and credal networks has been derived. This allows to reduce a counterfactual query in a causal model to a standard query in a credal network, even in the case of unidentifiable causal effects. The necessary transformations and data structures are implemented in CREDICI, while inferences are eventually computed by CREMA (Huber et al., 2020), a twin library for general credal network inference. Here we discuss the main implementation challenges and possible outlooks.

Cite this Paper


BibTeX
@InProceedings{pmlr-v138-cabanas20a, title = {CREDICI: A Java Library for Causal Inference by Credal Networks}, author = {Caba\~nas, Rafael and Antonucci, Alessandro and Huber, David and Zaffalon, Marco}, booktitle = {Proceedings of the 10th International Conference on Probabilistic Graphical Models}, pages = {597--600}, year = {2020}, editor = {Manfred Jaeger and Thomas Dyhre Nielsen}, volume = {138}, series = {Proceedings of Machine Learning Research}, month = {23--25 Sep}, publisher = {PMLR}, pdf = {http://proceedings.mlr.press/v138/cabanas20a/cabanas20a.pdf}, url = { http://proceedings.mlr.press/v138/cabanas20a.html }, abstract = {We present CREDICI, a Java open-source tool for causal inference based on credal networks. Credal networks are an extension of Bayesian networks where local probability mass functions are only constrained to belong to given, so-called credal, sets. CREDICI is based on the recent work of Zaffalon et al. (2020), where an equivalence between Pearl’s structural causal models and credal networks has been derived. This allows to reduce a counterfactual query in a causal model to a standard query in a credal network, even in the case of unidentifiable causal effects. The necessary transformations and data structures are implemented in CREDICI, while inferences are eventually computed by CREMA (Huber et al., 2020), a twin library for general credal network inference. Here we discuss the main implementation challenges and possible outlooks.} }
Endnote
%0 Conference Paper %T CREDICI: A Java Library for Causal Inference by Credal Networks %A Rafael Cabañas %A Alessandro Antonucci %A David Huber %A Marco Zaffalon %B Proceedings of the 10th International Conference on Probabilistic Graphical Models %C Proceedings of Machine Learning Research %D 2020 %E Manfred Jaeger %E Thomas Dyhre Nielsen %F pmlr-v138-cabanas20a %I PMLR %P 597--600 %U http://proceedings.mlr.press/v138/cabanas20a.html %V 138 %X We present CREDICI, a Java open-source tool for causal inference based on credal networks. Credal networks are an extension of Bayesian networks where local probability mass functions are only constrained to belong to given, so-called credal, sets. CREDICI is based on the recent work of Zaffalon et al. (2020), where an equivalence between Pearl’s structural causal models and credal networks has been derived. This allows to reduce a counterfactual query in a causal model to a standard query in a credal network, even in the case of unidentifiable causal effects. The necessary transformations and data structures are implemented in CREDICI, while inferences are eventually computed by CREMA (Huber et al., 2020), a twin library for general credal network inference. Here we discuss the main implementation challenges and possible outlooks.
APA
Cabañas, R., Antonucci, A., Huber, D. & Zaffalon, M.. (2020). CREDICI: A Java Library for Causal Inference by Credal Networks. Proceedings of the 10th International Conference on Probabilistic Graphical Models, in Proceedings of Machine Learning Research 138:597-600 Available from http://proceedings.mlr.press/v138/cabanas20a.html .

Related Material