research area Methods and Algorithms
content type paperpublished September 2025

Identifiable Multi-View Causal Discovery Without Non-Gaussianity

AuthorsAmbroise Heurtebise†*, Omar Chehab‡, Pierre Ablin, Alexandre Gramfort†, Aapo Hyvärinen§

View publication

We propose a novel approach to linear causal discovery in the framework of multi-view Structural Equation Models (SEM). Our proposed model relaxes the well-known assumption of non-Gaussian disturbances by alternatively assuming diversity of variances over views, making it more broadly applicable. We prove the identifiability of all the parameters of the model without any further assumptions on the structure of the SEM other than it being acyclic. We further propose an estimation algorithm based on recent advances in multi-view Independent Component Analysis (ICA). The proposed methodology is validated through simulations and application on real neuroimaging data, where it enables the estimation of causal graphs between brain regions.

  • † Inria, CEA, University of Paris-Saclay, France
  • ‡ ENSAE, CREST, IP Paris, France
  • § University of Helsinki, Finland
  • * Equal Contributor

Related readings and updates.

MVICAD2: Multi-View Independent Component Analysis with Delays and Dilations

August 18, 2026research area Methods and Algorithmsconference IEEE Transactions on Biomedical Engineering

Machine learning techniques in multi-view settings face significant challenges, particularly when integrating heterogeneous data, aligning feature spaces, and managing view-specific biases. These issues are prominent in neuroscience, where data from multiple subjects exposed to the same stimuli are analyzed to uncover brain activity dynamics. In magnetoencephalography (MEG), where signals are captured at the scalp level, estimating the brain’s…

Read more

Multi-View Causal Discovery without Non-Gaussianity: Identifiability and Algorithms

December 10, 2025research area Methods and Algorithms

Causal discovery is a difficult problem that typically relies on strong assumptions on the data-generating model, such as non-Gaussianity. In practice, many modern applications provide multiple related views of the same system, which has rarely been considered for causal discovery. Here, we leverage this multi-view structure to achieve causal discovery with weak assumptions. We propose a multi-view linear Structural Equation Model (SEM) that…

Read more