%0 Journal Article %1 goudet2017causal %A Goudet, Olivier %A Kalainathan, Diviyan %A Caillou, Philippe %A Guyon, Isabelle %A Lopez-Paz, David %A Sebag, Michèle %D 2017 %K causal-analysis generative-models %T Causal Generative Neural Networks %U http://arxiv.org/abs/1711.08936 %X We present Causal Generative Neural Networks (CGNNs) to learn functional causal models from observational data. CGNNs leverage conditional independencies and distributional asymmetries to discover bivariate and multivariate causal structures. CGNNs make no assumption regarding the lack of confounders, and learn a differentiable generative model of the data by using backpropagation. Extensive experiments show their good performances comparatively to the state of the art in observational causal discovery on both simulated and real data, with respect to cause-effect inference, v-structure identification, and multivariate causal discovery.

@article{goudet2017causal, abstract = {We present Causal Generative Neural Networks (CGNNs) to learn functional causal models from observational data. CGNNs leverage conditional independencies and distributional asymmetries to discover bivariate and multivariate causal structures. CGNNs make no assumption regarding the lack of confounders, and learn a differentiable generative model of the data by using backpropagation. Extensive experiments show their good performances comparatively to the state of the art in observational causal discovery on both simulated and real data, with respect to cause-effect inference, v-structure identification, and multivariate causal discovery.}, added-at = {2019-05-23T04:16:04.000+0200}, author = {Goudet, Olivier and Kalainathan, Diviyan and Caillou, Philippe and Guyon, Isabelle and Lopez-Paz, David and Sebag, Michèle}, biburl = {https://www.bibsonomy.org/bibtex/2e719903eae71d8b1456b458faaad2e4f/kirk86}, description = {[1711.08936] Causal Generative Neural Networks}, interhash = {0b4bc62d51532e800784a9e200efbd73}, intrahash = {e719903eae71d8b1456b458faaad2e4f}, keywords = {causal-analysis generative-models}, note = {cite arxiv:1711.08936}, timestamp = {2019-05-23T04:16:04.000+0200}, title = {Causal Generative Neural Networks}, url = {http://arxiv.org/abs/1711.08936}, year = 2017 }