Feedback System Neural Networks for Inferring Causality in Directed Cyclic Graphs

TL;DR

This paper introduces FSNN, a neural network-based causal inference method for directed cyclic graphs, using non-linear ODEs to model and interpret complex dynamic systems with proven accuracy.

Contribution

The paper presents a novel causal network learning algorithm (FSNN) that constructs interpretable non-linear ODE models using neural networks for dynamic systems.

Findings

Accurately models a three-state non-linear dynamic system

Provides causally correct insights into system behavior

Demonstrates rapid experimentation and analysis capabilities

Abstract

This paper presents a new causal network learning algorithm (FSNN, Feedback System Neural Network) based on the construction and analysis of a non-linear system of Ordinary Differential Equations (ODEs). The constructed system provides insight into the mechanisms responsible for generating the past and potential future behavior of dynamic systems. It is also interpretable in terms of real system variables, providing a wholistic, causally accurate, and systemic understanding of the real-life interactions governing observed phenomena. This paper demonstrates the generation of an n-dimensional ordinary differential equation model that can be parameterized to fit measured data using standard numerical optimization techniques. The model makes use of feed forward artificial neural nets to capture nonlinearity, but is a parsimonious and interpretable representation of the network of causal relationships in complex systems. The generated model can easily and rapidly be experimented with and analyzed to determine the origins of behavior using the loops that matter method (Schoenberg et. al 2019). A demonstration of the utility and applicability of the method is given, showing that it produces an accurate, and causally correct model for a three state, non-linear, complex dynamic system of known origin. Generalization to other dynamic systems with other data sources is then discussed.