Shadow-Mapping for Unsupervised Neural Causal Discovery

TL;DR

This paper introduces Neural Shadow-Mapping, a neural network approach that embeds high-dimensional video data into low-dimensional representations to uncover causal relationships in dynamic systems where traditional correlation-based methods fail.

Contribution

The paper presents a novel neural network method for causal discovery that effectively handles dynamic systems with mirage correlations, outperforming traditional correlation-based methods.

Findings

Successfully discovers causal links from video data of dynamic systems.

Handles 'mirage' correlations that vary with observation windows.

Outperforms existing correlation-based causal discovery methods.

Abstract

An important goal across most scientific fields is the discovery of causal structures underling a set of observations. Unfortunately, causal discovery methods which are based on correlation or mutual information can often fail to identify causal links in systems which exhibit dynamic relationships. Such dynamic systems (including the famous coupled logistic map) exhibit `mirage' correlations which appear and disappear depending on the observation window. This means not only that correlation is not causation but, perhaps counter-intuitively, that causation may occur without correlation. In this paper we describe Neural Shadow-Mapping, a neural network based method which embeds high-dimensional video data into a low-dimensional shadow representation, for subsequent estimation of causal links. We demonstrate its performance at discovering causal links from video-representations of dynamic systems.