Shadow-Mapping for Unsupervised Neural Causal Discovery
This addresses the challenge of identifying causal structures in dynamic systems for researchers in fields like physics or biology, though it appears incremental as it builds on neural network methods for a known bottleneck.
The paper tackled the problem of causal discovery in dynamic systems where traditional correlation-based methods fail due to 'mirage' correlations, and introduced Neural Shadow-Mapping to embed high-dimensional video data into low-dimensional representations for estimating causal links, demonstrating its performance on video data of dynamic systems.
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.