Discovering dependencies in complex physical systems using Neural Networks

TL;DR

This paper introduces a neural network-based method leveraging mutual information to uncover complex, non-linear dependencies and causal relationships in multivariable dynamical systems, effective even with limited data.

Contribution

The paper presents a novel neural network framework that detects non-linear dependencies and causality in complex systems, applicable to small datasets.

Findings

Effective in identifying non-linear relationships

Works with limited data points

Applicable to real-world dynamical systems

Abstract

In todays age of data, discovering relationships between different variables is an interesting and a challenging problem. This problem becomes even more critical with regards to complex dynamical systems like weather forecasting and econometric models, which can show highly non-linear behavior. A method based on mutual information and deep neural networks is proposed as a versatile framework for discovering non-linear relationships ranging from functional dependencies to causality. We demonstrate the application of this method to actual multivariable non-linear dynamical systems. We also show that this method can find relationships even for datasets with small number of datapoints, as is often the case with empirical data.