Group Interventions on Deep Networks for Causal Discovery in Subsystems
This work addresses the limitation of existing causal discovery methods that overlook group interactions, offering a novel approach for domains such as neuroscience and climate science, though it is incremental in extending deep learning techniques to group-level analysis.
The paper tackled the problem of causal discovery in nonlinear multivariate time series by focusing on group-level interactions rather than pairwise relationships, introducing gCDMI, a method that uses group interventions on deep neural networks and invariance testing, and demonstrated superior performance in identifying group-level causal links on simulated and real-world datasets like brain networks and climate ecosystems.
Causal discovery uncovers complex relationships between variables, enhancing predictions, decision-making, and insights into real-world systems, especially in nonlinear multivariate time series. However, most existing methods primarily focus on pairwise cause-effect relationships, overlooking interactions among groups of variables, i.e., subsystems and their collective causal influence. In this study, we introduce gCDMI, a novel multi-group causal discovery method that leverages group-level interventions on trained deep neural networks and employs model invariance testing to infer causal relationships. Our approach involves three key steps. First, we use deep learning to jointly model the structural relationships among groups of all time series. Second, we apply group-wise interventions to the trained model. Finally, we conduct model invariance testing to determine the presence of causal links among variable groups. We evaluate our method on simulated datasets, demonstrating its superior performance in identifying group-level causal relationships compared to existing methods. Additionally, we validate our approach on real-world datasets, including brain networks and climate ecosystems. Our results highlight that applying group-level interventions to deep learning models, combined with invariance testing, can effectively reveal complex causal structures, offering valuable insights for domains such as neuroscience and climate science.