Investigating Sindy As a Tool For Causal Discovery In Time Series Signals
This work addresses the need for engineers to identify robust causal relationships in dynamical systems, but it is incremental as it builds on existing SINDy and causal discovery methods.
The paper tackles the problem of causal discovery in time series signals by proposing to augment the SINDy algorithm with causal discovery tools, resulting in improved performance for learning causally robust governing equations, as demonstrated empirically.
The SINDy algorithm has been successfully used to identify the governing equations of dynamical systems from time series data. In this paper, we argue that this makes SINDy a potentially useful tool for causal discovery and that existing tools for causal discovery can be used to dramatically improve the performance of SINDy as tool for robust sparse modeling and system identification. We then demonstrate empirically that augmenting the SINDy algorithm with tools from causal discovery can provides engineers with a tool for learning causally robust governing equations.