Back
Telling cause from effect based on high-dimensional observations
We describe a method for inferring linear causal relations among multi-dimensional variables. The idea is to use an asymmetry between the distributions of cause and effect that occurs if the covariance matrix of the cause and the structure matrix mapping the cause to the effect are independently chosen. The method applies to both stochastic and deterministic causal relations, provided that the dimensionality is sufficiently high (in some experiments, 5 was enough). It is applicable to Gaussian as well as non-Gaussian data.
@inproceedings{6501, title = {Telling cause from effect based on high-dimensional observations}, booktitle = {Proceedings of the 27th International Conference on Machine Learning}, abstract = {We describe a method for inferring linear causal relations among multi-dimensional variables. The idea is to use an asymmetry between the distributions of cause and effect that occurs if the covariance matrix of the cause and the structure matrix mapping the cause to the effect are independently chosen. The method applies to both stochastic and deterministic causal relations, provided that the dimensionality is sufficiently high (in some experiments, 5 was enough). It is applicable to Gaussian as well as non-Gaussian data.}, pages = {479-486}, editors = {J F{\"u}rnkranz and T Joachims}, publisher = {International Machine Learning Society}, organization = {Max-Planck-Gesellschaft}, school = {Biologische Kybernetik}, address = {Madison, WI, USA}, month = jun, year = {2010}, slug = {6501}, author = {Janzing, D. and Hoyer, P. and Sch{\"o}lkopf, B.}, month_numeric = {6} }