Causal Network Inference by Optimal Causation Entropy

Abstract

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 forms 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 to develop 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-couple dynamical system. Validation in terms of analytical and numerical results for Gaussian processes on large random networks highlight that inference by our algorithm outperforms previous leading methods including conditioned 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.

References in corpus (4)

• ARACNE: An Algorithm for the Reconstruction of Gene Regulatory Networks in a Mammalian Cellular Context
• Critical phenomena in complex networks
• Causation Entropy Identifies Indirect Influences, Dominance of Neighbors and Anticipatory Couplings
• Robustness of Optimal Synchronization in Real Networks

Cited by in corpus (53)

• Detecting causal associations in large nonlinear time series datasets
• Coupling functions: Universal insights into dynamical interaction mechanisms
• Data Based Identification and Prediction of Nonlinear and Complex Dynamical Systems
• On Explaining the Surprising Success of Reservoir Computing Forecaster of Chaos? The Universal Machine Learning Dynamical System with Contrasts to VAR and DMD
• Network reconstruction and community detection from dynamics
• Large-scale directed network inference with multivariate transfer entropy and hierarchical statistical testing
• Autonomous inference of complex network dynamics from incomplete and noisy data
• Inferring network properties from time series using transfer entropy and mutual information: validation of multivariate versus bivariate approaches
• How Entropic Regression Beats the Outliers Problem in Nonlinear System Identification
• Inference of Causal Information Flow in Collective Animal Behavior
• Identifying Coupling Structure in Complex Systems through the Optimal Causation Entropy Principle
• Understanding causation via correlations and linear response theory
• Anatomy of Leadership in Collective Behaviour
• Causation entropy from symbolic representations of dynamical systems
• Hypergraph reconstruction from dynamics
• Geometric k-nearest neighbor estimation of entropy and mutual information
• Detecting causality in multivariate time series via non-uniform embedding
• Reconstructing directed and weighted topologies of phase-locked oscillator networks
• Machine Learning Link Inference of Noisy Delay-coupled Networks with Opto-Electronic Experimental Tests
• Deriving pairwise transfer entropy from network structure and motifs
• Modes of Information Flow in Collective Cohesion
• Measuring topological descriptors of complex networks under uncertainty
• Causal Inference for Time series Analysis: Problems, Methods and Evaluation
• Complex contagion features without social reinforcement in a model of social information flow
• Connecting Network Science and Information Theory
• Sensitive Dependence of Optimal Network Dynamics on Network Structure
• Information theoretical noninvasive damage detection in bridge structures
• Entropic Causal Inference for Neurological Applications
• Extracting causation from millennial-scale climate fluctuations in the last 800 kyr
• CEBoosting: Online Sparse Identification of Dynamical Systems with Regime Switching by Causation Entropy Boosting
• Response and flux of information in extended non-equilibrium dynamics
• Transfer entropy dependent on distance among agents in quantifying leader-follower relationships
• Open or Closed? Information Flow Decided by Transfer Operators and Forecastability Quality Metric
• A Framework for Causal Discovery in non-intervenable systems
• Network reconstruction via the minimum description length principle
• Causal network discovery by iterative conditioning: comparison of algorithms
• Inference of topology and the nature of synapses, and the flow of information in neuronal networks
• The quoter model: a paradigmatic model of the social flow of written information
• The Many Faces of Fluctuation-Dissipation Relations Out of Equilibrium
• Towards a Framework for Observational Causality From Time Series: When Shannon Meets Turing
• On Geometry of Information Flow for Causal Inference
• A Rigorous Information-Theoretic Definition of Redundancy and Relevancy in Feature Selection Based on (Partial) Information Decomposition
• The Design of Global Correlation Quantifiers and Continuous Notions of Statistical Sufficiency
• Causal Link Discovery with Unequal Edge Error Tolerance
• Lower Bounds on Information Requirements for Causal Network Inference
• Mixing Times and Structural Inference for Bernoulli Autoregressive Processes
• Entropic Regression DMD (ERDMD) Discovers Informative Sparse and Nonuniformly Time Delayed Models
• Assimilative Causal Inference
• Small-worldness favours network inference
• Inference of Stochastic Dynamical Systems from Cross-Sectional Population Data
• Data-Driven Learning of Boolean Networks and Functions by Optimal Causation Entropy Principle (BoCSE)
• Causal inference in time series in terms of Rényi transfer entropy
• Spatio-Temporal Graphical Counterfactuals: An Overview