We gratefully acknowledge support from
the Simons Foundation and member institutions.
Full-text links:

Download:

Current browse context:

cs.IT

Change to browse by:

References & Citations

DBLP - CS Bibliography

Bookmark

(what is this?)
CiteULike logo BibSonomy logo Mendeley logo del.icio.us logo Digg logo Reddit logo ScienceWISE logo

Computer Science > Information Theory

Title: Lower Bounds on Information Requirements for Causal Network Inference

Abstract: Recovery of the causal structure of dynamic networks from noisy measurements has long been a problem of intense interest across many areas of science and engineering. Many algorithms have been proposed, but there is no work that compares the performance of the algorithms to converse bounds in a non-asymptotic setting. As a step to address this problem, this paper gives lower bounds on the error probability for causal network support recovery in a linear Gaussian setting. The bounds are based on the use of the Bhattacharyya coefficient for binary hypothesis testing problems with mixture probability distributions. Comparison of the bounds and the performance achieved by two representative recovery algorithms are given for sparse random networks based on the Erd\H{o}s-R\'enyi model.
Comments: Full version of the ISIT paper
Subjects: Information Theory (cs.IT)
Cite as: arXiv:2102.00055 [cs.IT]
  (or arXiv:2102.00055v2 [cs.IT] for this version)

Submission history

From: Xiaohan Kang [view email]
[v1] Fri, 29 Jan 2021 20:34:28 GMT (86kb)
[v2] Fri, 7 May 2021 14:44:12 GMT (145kb)

Link back to: arXiv, form interface, contact.