SADA: A General Framework to Support Robust Causation Discovery

Ruichu Cai, Zhenjie Zhang, Zhifeng Hao
Proceedings of the 30th International Conference on Machine Learning, PMLR 28(2):208-216, 2013.

Abstract

Causality discovery without manipulation is considered a crucial problem to a variety of applications, such as genetic therapy. The state-of-the-art solutions, e.g. LiNGAM, return accurate results when the number of labeled samples is larger than the number of variables. These approaches are thus applicable only when large numbers of samples are available or the problem domain is sufficiently small. Motivated by the observations of the local sparsity properties on causal structures, we propose a general Split-and-Merge strategy, named SADA, to enhance the scalability of a wide class of causality discovery algorithms. SADA is able to accurately identify the causal variables, even when the sample size is significantly smaller than the number of variables. In SADA, the variables are partitioned into subsets, by finding cuts on the sparse probabilistic graphical model over the variables. By running mainstream causation discovery algorithms, e.g. LiNGAM, on the subproblems, complete causality can be reconstructed by combining all the partial results. SADA benefits from the recursive division technique, since each small subproblem generates more accurate result under the same number of samples. We theoretically prove that SADA always reduces the scale of problems without significant sacrifice on result accuracy, depending only on the local sparsity condition over the variables. Experiments on real-world datasets verify the improvements on scalability and accuracy by applying SADA on top of existing causation algorithms.

Cite this Paper

BibTeX
@InProceedings{pmlr-v28-cai13, title = {SADA: A General Framework to Support Robust Causation Discovery}, author = {Cai, Ruichu and Zhang, Zhenjie and Hao, Zhifeng}, booktitle = {Proceedings of the 30th International Conference on Machine Learning}, pages = {208--216}, year = {2013}, editor = {Dasgupta, Sanjoy and McAllester, David}, volume = {28}, number = {2}, series = {Proceedings of Machine Learning Research}, address = {Atlanta, Georgia, USA}, month = {17--19 Jun}, publisher = {PMLR}, pdf = {http://proceedings.mlr.press/v28/cai13.pdf}, url = {https://proceedings.mlr.press/v28/cai13.html}, abstract = {Causality discovery without manipulation is considered a crucial problem to a variety of applications, such as genetic therapy. The state-of-the-art solutions, e.g. LiNGAM, return accurate results when the number of labeled samples is larger than the number of variables. These approaches are thus applicable only when large numbers of samples are available or the problem domain is sufficiently small. Motivated by the observations of the local sparsity properties on causal structures, we propose a general Split-and-Merge strategy, named SADA, to enhance the scalability of a wide class of causality discovery algorithms. SADA is able to accurately identify the causal variables, even when the sample size is significantly smaller than the number of variables. In SADA, the variables are partitioned into subsets, by finding cuts on the sparse probabilistic graphical model over the variables. By running mainstream causation discovery algorithms, e.g. LiNGAM, on the subproblems, complete causality can be reconstructed by combining all the partial results. SADA benefits from the recursive division technique, since each small subproblem generates more accurate result under the same number of samples. We theoretically prove that SADA always reduces the scale of problems without significant sacrifice on result accuracy, depending only on the local sparsity condition over the variables. Experiments on real-world datasets verify the improvements on scalability and accuracy by applying SADA on top of existing causation algorithms.} }
Endnote
%0 Conference Paper %T SADA: A General Framework to Support Robust Causation Discovery %A Ruichu Cai %A Zhenjie Zhang %A Zhifeng Hao %B Proceedings of the 30th International Conference on Machine Learning %C Proceedings of Machine Learning Research %D 2013 %E Sanjoy Dasgupta %E David McAllester %F pmlr-v28-cai13 %I PMLR %P 208--216 %U https://proceedings.mlr.press/v28/cai13.html %V 28 %N 2 %X Causality discovery without manipulation is considered a crucial problem to a variety of applications, such as genetic therapy. The state-of-the-art solutions, e.g. LiNGAM, return accurate results when the number of labeled samples is larger than the number of variables. These approaches are thus applicable only when large numbers of samples are available or the problem domain is sufficiently small. Motivated by the observations of the local sparsity properties on causal structures, we propose a general Split-and-Merge strategy, named SADA, to enhance the scalability of a wide class of causality discovery algorithms. SADA is able to accurately identify the causal variables, even when the sample size is significantly smaller than the number of variables. In SADA, the variables are partitioned into subsets, by finding cuts on the sparse probabilistic graphical model over the variables. By running mainstream causation discovery algorithms, e.g. LiNGAM, on the subproblems, complete causality can be reconstructed by combining all the partial results. SADA benefits from the recursive division technique, since each small subproblem generates more accurate result under the same number of samples. We theoretically prove that SADA always reduces the scale of problems without significant sacrifice on result accuracy, depending only on the local sparsity condition over the variables. Experiments on real-world datasets verify the improvements on scalability and accuracy by applying SADA on top of existing causation algorithms.
RIS
TY - CPAPER TI - SADA: A General Framework to Support Robust Causation Discovery AU - Ruichu Cai AU - Zhenjie Zhang AU - Zhifeng Hao BT - Proceedings of the 30th International Conference on Machine Learning DA - 2013/05/13 ED - Sanjoy Dasgupta ED - David McAllester ID - pmlr-v28-cai13 PB - PMLR DP - Proceedings of Machine Learning Research VL - 28 IS - 2 SP - 208 EP - 216 L1 - http://proceedings.mlr.press/v28/cai13.pdf UR - https://proceedings.mlr.press/v28/cai13.html AB - Causality discovery without manipulation is considered a crucial problem to a variety of applications, such as genetic therapy. The state-of-the-art solutions, e.g. LiNGAM, return accurate results when the number of labeled samples is larger than the number of variables. These approaches are thus applicable only when large numbers of samples are available or the problem domain is sufficiently small. Motivated by the observations of the local sparsity properties on causal structures, we propose a general Split-and-Merge strategy, named SADA, to enhance the scalability of a wide class of causality discovery algorithms. SADA is able to accurately identify the causal variables, even when the sample size is significantly smaller than the number of variables. In SADA, the variables are partitioned into subsets, by finding cuts on the sparse probabilistic graphical model over the variables. By running mainstream causation discovery algorithms, e.g. LiNGAM, on the subproblems, complete causality can be reconstructed by combining all the partial results. SADA benefits from the recursive division technique, since each small subproblem generates more accurate result under the same number of samples. We theoretically prove that SADA always reduces the scale of problems without significant sacrifice on result accuracy, depending only on the local sparsity condition over the variables. Experiments on real-world datasets verify the improvements on scalability and accuracy by applying SADA on top of existing causation algorithms. ER -
APA
Cai, R., Zhang, Z. & Hao, Z.. (2013). SADA: A General Framework to Support Robust Causation Discovery. Proceedings of the 30th International Conference on Machine Learning, in Proceedings of Machine Learning Research 28(2):208-216 Available from https://proceedings.mlr.press/v28/cai13.html.

Related Material