Causal Discovery on Higher-Order Interactions
By: Alessio Zanga, Marco Scutari, Fabio Stella
Potential Business Impact:
Finds hidden causes in data, even with little information.
Causal discovery combines data with knowledge provided by experts to learn the DAG representing the causal relationships between a given set of variables. When data are scarce, bagging is used to measure our confidence in an average DAG obtained by aggregating bootstrapped DAGs. However, the aggregation step has received little attention from the specialized literature: the average DAG is constructed using only the confidence in the individual edges of the bootstrapped DAGs, thus disregarding complex higher-order edge structures. In this paper, we introduce a novel theoretical framework based on higher-order structures and describe a new DAG aggregation algorithm. We perform a simulation study, discussing the advantages and limitations of the proposed approach. Our proposal is both computationally efficient and effective, outperforming state-of-the-art solutions, especially in low sample size regimes and under high dimensionality settings.
Similar Papers
Causal Discovery for Linear DAGs with Dependent Latent Variables via Higher-order Cumulants
Machine Learning (CS)
Finds hidden causes in data, even when they're tricky.
Higher-Order Causal Structure Learning with Additive Models
Machine Learning (CS)
Finds hidden connections that affect how things work.
Causal DAG Summarization (Full Version)
Machine Learning (CS)
Simplifies complex cause-and-effect maps for easier study.