Your access to this SIAM content is provided through the subscription of University of Bath

SIAM Journal on Applied Dynamical Systems


Volume 14, Issue 1

Causal Network Inference by Optimal Causation Entropy

Related Databases

Web of Science

You must be logged in with an active subscription to view this.

Article Data

History

Submitted: 10 February 2014
Accepted: 11 November 2014
Published online: 08 January 2015

Publication Data

ISSN (online): 1536-0040
CODEN: sjaday

The broad abundance of time series data, which is in sharp contrast to limited knowledge of the underlying network dynamic processes that produce such observations, calls for a rigorous and efficient method of causal network inference. Here we develop mathematical theory of causation entropy, an information-theoretic statistic designed for model-free causality inference. For stationary Markov processes, we prove that for a given node in the network, its causal parents form the minimal set of nodes that maximizes causation entropy, a result we refer to as the optimal causation entropy principle. Furthermore, this principle guides us in developing computational and data efficient algorithms for causal network inference based on a two-step discovery and removal algorithm for time series data for a network-coupled dynamical system. Validation in terms of analytical and numerical results for Gaussian processes on large random networks highlights that inference by our algorithm outperforms previous leading methods, including conditional Granger causality and transfer entropy. Interestingly, our numerical results suggest that the number of samples required for accurate inference depends strongly on network characteristics such as the density of links and information diffusion rate and not necessarily on the number of nodes.

© 2015, Society for Industrial and Applied Mathematics

Cited by

(2019) Two stage approach to functional network reconstruction for binary time-series. The European Physical Journal B 92:2. Crossref
2019. Inferring Causal Gene Regulatory Networks Using Time-Delay Association Rules. Advanced Informatics for Computing Research, 310-321. Crossref
(2018) Detecting intermittent switching leadership in coupled dynamical systems. Scientific Reports 8:1. Crossref
(2018) Connecting network science and information theory. Physica A: Statistical Mechanics and its Applications. Crossref
(2018) The quoter model: A paradigmatic model of the social flow of written information. Chaos: An Interdisciplinary Journal of Nonlinear Science 28:7, 075304. Crossref
(2018) Causality, dynamical systems and the arrow of time. Chaos: An Interdisciplinary Journal of Nonlinear Science 28:7, 075307. Crossref
(2018) Open or closed? Information flow decided by transfer operators and forecastability quality metric. Chaos: An Interdisciplinary Journal of Nonlinear Science 28:7, 075309. Crossref
(2018) Anatomy of leadership in collective behaviour. Chaos: An Interdisciplinary Journal of Nonlinear Science 28:7, 075308. Crossref
(2018) Introduction to Focus Issue: Causation inference and information flow in dynamical systems: Theory and applications. Chaos: An Interdisciplinary Journal of Nonlinear Science 28:7, 075201. Crossref
(2018) Causal inference for multivariate stochastic process prediction. Information Sciences 448-449, 134-148. Crossref
(2018) Information Theoretic Causality Measures for System Identification of Mechanical Systems. Journal of Computational and Nonlinear Dynamics 13:7, 071005. Crossref
(2018) Inferring influence and leadership in moving animal groups. Philosophical Transactions of the Royal Society B: Biological Sciences 373:1746, 20170006. Crossref
(2018) Geometric k-nearest neighbor estimation of entropy and mutual information. Chaos: An Interdisciplinary Journal of Nonlinear Science 28:3, 033114. Crossref
(2018) Inference of topology and the nature of synapses, and the flow of information in neuronal networks. Physical Review E 97:2. Crossref
(2018) Causal network reconstruction from time series: From theoretical assumptions to practical estimation. Chaos: An Interdisciplinary Journal of Nonlinear Science 28:7, 075310. Crossref
(2017) Coupling functions: Universal insights into dynamical interaction mechanisms. Reviews of Modern Physics 89:4. Crossref
(2017) Sensitive Dependence of Optimal Network Dynamics on Network Structure. Physical Review X 7:4. Crossref
(2017) Potential conditional mutual information: Estimators and properties. 2017 55th Annual Allerton Conference on Communication, Control, and Computing (Allerton), 1228-1235. Crossref
(2017) Do causal concentration–response functions exist? A critical review of associational and causal relations between fine particulate matter and mortality. Critical Reviews in Toxicology 47:7, 609-637. Crossref
(2017) Detection of coupling delay: A problem not yet solved. Chaos: An Interdisciplinary Journal of Nonlinear Science 27:8, 083109. Crossref
(2017) Detection of time delays and directional interactions based on time series from complex dynamical systems. Physical Review E 96:1. Crossref
(2017) Reconstructing the structure of directed and weighted networks of nonlinear oscillators. Physical Review E 95:4. Crossref
(2017) Information flow in a model of policy diffusion: an analytical study. IEEE Transactions on Network Science and Engineering, 1-1. Crossref
(2017) Causation Entropy Identifies Sparsity Structure for Parameter Estimation of Dynamic Systems. Journal of Computational and Nonlinear Dynamics 12:1, 011008. Crossref
(2016) Information-theoretical noninvasive damage detection in bridge structures. Chaos: An Interdisciplinary Journal of Nonlinear Science 26:11, 116312. Crossref
(2016) BOLD fMRI complexity predicts changes in brain processes, interactions and patterns, in health and disease. Journal of the Neurological Sciences 367, 347-348. Crossref
(2016) Data based identification and prediction of nonlinear and complex dynamical systems. Physics Reports 644, 1-76. Crossref
(2016) Directed dynamical influence is more detectable with noise. Scientific Reports 6:1. Crossref
(2016) Inference of Causal Information Flow in Collective Animal Behavior. IEEE Transactions on Molecular, Biological and Multi-Scale Communications 2:1, 107-116. Crossref
, , , , , and . (2016) Understanding Policy Diffusion in the U.S.: An Information-Theoretical Approach to Unveil Connectivity Structures in Slowly Evolving Complex Systems. SIAM Journal on Applied Dynamical Systems 15:3, 1384-1409. Abstract | PDF (1175 KB) 
(2015) Causation entropy from symbolic representations of dynamical systems. Chaos: An Interdisciplinary Journal of Nonlinear Science 25:4, 043106. Crossref