Permutation-based Causal Inference Algorithms with Interventions
Name
7164-permutation-based-causal-inference-algorithms-with-interventions.pdf
Description
Published version
Size
626.33 KB
Format
Adobe PDF
Checksum (MD5)
34a2a53bcf06a5b2653993adc01979e0
Author(s)
Wang, Yuhao
Solus, Liam
Yang, Karren Dai
Uhler, Caroline
Date Issued
2017
Citation
Wang, Yuhao, Solus, Liam, Yang, Karren Dai and Uhler, Caroline. 2017. "Permutation-based Causal Inference Algorithms with Interventions."
Version
Final published version
Abstract
© 2017 Neural information processing systems foundation. All rights reserved. Learning directed acyclic graphs using both observational and interventional data is now a fundamentally important problem due to recent technological developments in genomics that generate such single-cell gene expression data at a very large scale. In order to utilize this data for learning gene regulatory networks, efficient and reliable causal inference algorithms are needed that can make use of both observational and interventional data. In this paper, we present two algorithms of this type and prove that both are consistent under the faithfulness assumption. These algorithms are interventional adaptations of the Greedy SP algorithm and are the first algorithms using both observational and interventional data with consistency guarantees. Moreover, these algorithms have the advantage that they are nonparametric, which makes them useful also for analyzing non-Gaussian data. In this paper, we present these two algorithms and their consistency guarantees, and we analyze their performance on simulated data, protein signaling data, and single-cell gene expression data.
MIT Department
Massachusetts Institute of Technology. Laboratory for Information and Decision Systems
Massachusetts Institute of Technology. Institute for Data, Systems, and Society
Terms of Use
Article is made available in accordance with the publisher's policy and may be subject to US copyright law. Please refer to the publisher's site for terms of use.
Persistent DSpace Link
