Deep Learning of Causal Structures in High Dimensions
This addresses the challenge of causal inference in large-scale scientific applications, particularly in biomedicine, by providing a scalable deep learning approach.
The paper tackles the problem of learning causal relationships in high-dimensional data, such as in biomedicine, by proposing a deep neural architecture that combines convolutional and graph neural networks within a causal risk framework, demonstrating feasibility on simulations and real biological data with thousands of variables.
Recent years have seen rapid progress at the intersection between causality and machine learning. Motivated by scientific applications involving high-dimensional data, in particular in biomedicine, we propose a deep neural architecture for learning causal relationships between variables from a combination of empirical data and prior causal knowledge. We combine convolutional and graph neural networks within a causal risk framework to provide a flexible and scalable approach. Empirical results include linear and nonlinear simulations (where the underlying causal structures are known and can be directly compared against), as well as a real biological example where the models are applied to high-dimensional molecular data and their output compared against entirely unseen validation experiments. These results demonstrate the feasibility of using deep learning approaches to learn causal networks in large-scale problems spanning thousands of variables.