CaltechAUTHORS
  A Caltech Library Service

Stability of Causal Inference

Schulman, Leonard J. and Srivastava, Piyush (2016) Stability of Causal Inference. In: UAI 2016 - Proceedings. , Art. No. 214. http://resolver.caltech.edu/CaltechAUTHORS:20160628-152001110

[img] PDF - Published Version
See Usage Policy.

292Kb

Use this Persistent URL to link to this item: http://resolver.caltech.edu/CaltechAUTHORS:20160628-152001110

Abstract

We consider the sensitivity of causal identification to small perturbations in the input. A long line of work culminating in papers by Shpitser and Pearl (2006) and Huang and Valtorta (2008) led to a complete procedure for the causal identification problem. In our main result in this paper, we show that the identification function computed by these procedures is in some cases extremely unstable numerically. Specifically, the “condition number” of causal identification can be of the order of Ω(exp(n ^(0.49))) on an identifiable semiMarkovian model with n visible nodes. That is, in order to give an output accurate to d bits, the empirical probabilities of the observable events need to be obtained to accuracy d + Ω(n ^(0.49)) bits.


Item Type:Book Section
Related URLs:
URLURL TypeDescription
http://auai.org/uai2016/proceedings.phpPublisherTable of Contents
http://auai.org/uai2016/proceedings/papers/214.pdfPublisherArticle
Additional Information:This research was supported by NSF grant CCF-1319745. We thank the reviewers for helpful comments.
Funders:
Funding AgencyGrant Number
NSFCCF-1319745
Record Number:CaltechAUTHORS:20160628-152001110
Persistent URL:http://resolver.caltech.edu/CaltechAUTHORS:20160628-152001110
Usage Policy:No commercial reproduction, distribution, display or performance rights in this work are provided.
ID Code:68719
Collection:CaltechAUTHORS
Deposited By: Joy Painter
Deposited On:28 Jun 2016 22:40
Last Modified:30 Jun 2016 17:07

Repository Staff Only: item control page