Combining Experiments to Discover Linear Cyclic Models with Latent Variables

Frederick Eberhardt, Patrik Hoyer, Richard Scheines
Proceedings of the Thirteenth International Conference on Artificial Intelligence and Statistics, PMLR 9:185-192, 2010.

Abstract

We present an algorithm to infer causal relations between a set of measured variables on the basis of experiments on these variables. The algorithm assumes that the causal relations are linear, but is otherwise completely general: It provides consistent estimates when the true causal structure contains feedback loops and latent variables, while the experiments can involve surgical or ’soft’ interventions on one or multiple variables at a time. The algorithm is ’online’ in the sense that it combines the results from any set of available experiments, can incorporate background knowledge and resolves conflicts that arise from combining results from different experiments. In addition we provide a necessary and sufficient condition that (i) determines when the algorithm can uniquely return the true graph, and (ii) can be used to select the next best experiment until this condition is satisfied. We demonstrate the method by applying it to simulated data and the flow cytometry data of Sachs et al (2005).

Cite this Paper

BibTeX
@InProceedings{pmlr-v9-eberhardt10a, title = {Combining Experiments to Discover Linear Cyclic Models with Latent Variables}, author = {Eberhardt, Frederick and Hoyer, Patrik and Scheines, Richard}, booktitle = {Proceedings of the Thirteenth International Conference on Artificial Intelligence and Statistics}, pages = {185--192}, year = {2010}, editor = {Teh, Yee Whye and Titterington, Mike}, volume = {9}, series = {Proceedings of Machine Learning Research}, address = {Chia Laguna Resort, Sardinia, Italy}, month = {13--15 May}, publisher = {PMLR}, pdf = {http://proceedings.mlr.press/v9/eberhardt10a/eberhardt10a.pdf}, url = {https://proceedings.mlr.press/v9/eberhardt10a.html}, abstract = {We present an algorithm to infer causal relations between a set of measured variables on the basis of experiments on these variables. The algorithm assumes that the causal relations are linear, but is otherwise completely general: It provides consistent estimates when the true causal structure contains feedback loops and latent variables, while the experiments can involve surgical or ’soft’ interventions on one or multiple variables at a time. The algorithm is ’online’ in the sense that it combines the results from any set of available experiments, can incorporate background knowledge and resolves conflicts that arise from combining results from different experiments. In addition we provide a necessary and sufficient condition that (i) determines when the algorithm can uniquely return the true graph, and (ii) can be used to select the next best experiment until this condition is satisfied. We demonstrate the method by applying it to simulated data and the flow cytometry data of Sachs et al (2005).} }
Endnote
%0 Conference Paper %T Combining Experiments to Discover Linear Cyclic Models with Latent Variables %A Frederick Eberhardt %A Patrik Hoyer %A Richard Scheines %B Proceedings of the Thirteenth International Conference on Artificial Intelligence and Statistics %C Proceedings of Machine Learning Research %D 2010 %E Yee Whye Teh %E Mike Titterington %F pmlr-v9-eberhardt10a %I PMLR %P 185--192 %U https://proceedings.mlr.press/v9/eberhardt10a.html %V 9 %X We present an algorithm to infer causal relations between a set of measured variables on the basis of experiments on these variables. The algorithm assumes that the causal relations are linear, but is otherwise completely general: It provides consistent estimates when the true causal structure contains feedback loops and latent variables, while the experiments can involve surgical or ’soft’ interventions on one or multiple variables at a time. The algorithm is ’online’ in the sense that it combines the results from any set of available experiments, can incorporate background knowledge and resolves conflicts that arise from combining results from different experiments. In addition we provide a necessary and sufficient condition that (i) determines when the algorithm can uniquely return the true graph, and (ii) can be used to select the next best experiment until this condition is satisfied. We demonstrate the method by applying it to simulated data and the flow cytometry data of Sachs et al (2005).
RIS
TY - CPAPER TI - Combining Experiments to Discover Linear Cyclic Models with Latent Variables AU - Frederick Eberhardt AU - Patrik Hoyer AU - Richard Scheines BT - Proceedings of the Thirteenth International Conference on Artificial Intelligence and Statistics DA - 2010/03/31 ED - Yee Whye Teh ED - Mike Titterington ID - pmlr-v9-eberhardt10a PB - PMLR DP - Proceedings of Machine Learning Research VL - 9 SP - 185 EP - 192 L1 - http://proceedings.mlr.press/v9/eberhardt10a/eberhardt10a.pdf UR - https://proceedings.mlr.press/v9/eberhardt10a.html AB - We present an algorithm to infer causal relations between a set of measured variables on the basis of experiments on these variables. The algorithm assumes that the causal relations are linear, but is otherwise completely general: It provides consistent estimates when the true causal structure contains feedback loops and latent variables, while the experiments can involve surgical or ’soft’ interventions on one or multiple variables at a time. The algorithm is ’online’ in the sense that it combines the results from any set of available experiments, can incorporate background knowledge and resolves conflicts that arise from combining results from different experiments. In addition we provide a necessary and sufficient condition that (i) determines when the algorithm can uniquely return the true graph, and (ii) can be used to select the next best experiment until this condition is satisfied. We demonstrate the method by applying it to simulated data and the flow cytometry data of Sachs et al (2005). ER -
APA
Eberhardt, F., Hoyer, P. & Scheines, R.. (2010). Combining Experiments to Discover Linear Cyclic Models with Latent Variables. Proceedings of the Thirteenth International Conference on Artificial Intelligence and Statistics, in Proceedings of Machine Learning Research 9:185-192 Available from https://proceedings.mlr.press/v9/eberhardt10a.html.

Related Material