Permutation-based Causal Inference Algorithms with Interventions

Abstract

Learning directed acyclic graphs using both observational and interventional data is now a fundamentally important problem due to recent technological developments in genomics that generate such single-cell gene expression data at a very large scale. In order to utilize this data for learning gene regulatory networks, efficient and reliable causal inference algorithms are needed that can make use of both observational and interventional data. In this paper, we present two algorithms of this type and prove that both are consistent under the faithfulness assumption. These algorithms are interventional adaptations of the Greedy SP algorithm and are the first algorithms using both observational and interventional data with consistency guarantees. Moreover, these algorithms have the advantage that they are nonparametric, which makes them useful also for analyzing non-Gaussian data. In this paper, we present these two algorithms and their consistency guarantees, and we analyze their performance on simulated data, protein signaling data, and single-cell gene expression data.

References in corpus (3)

• A Transformational Characterization of Equivalent Bayesian Network Structures
• Constraint-based Causal Discovery from Multiple Interventions over Overlapping Variable Sets
• Ancestral Causal Inference

Cited by in corpus (17)

• Causal Network Models of SARS-CoV-2 Expression and Aging to Identify Candidates for Drug Repurposing
• Graphical Criteria for Efficient Total Effect Estimation via Adjustment in Causal Linear Models
• Causal Discovery in Physical Systems from Videos
• Learning and Testing Causal Models with Interventions
• Differentiable Causal Discovery from Interventional Data
• Efficient adjustment sets for population average treatment effect estimation in non-parametric causal graphical models
• ABCD-Strategy: Budgeted Experimental Design for Targeted Causal Structure Discovery
• Efficient least squares for estimating total effects under linearity and causal sufficiency
• Permutation-Based Causal Structure Learning with Unknown Intervention Targets
• Near-Optimal Multi-Perturbation Experimental Design for Causal Structure Learning
• Identifying causal effects in maximally oriented partially directed acyclic graphs
• Efficient Neural Causal Discovery without Acyclicity Constraints
• Variance Minimization in the Wasserstein Space for Invariant Causal Prediction
• Explaining the Behavior of Black-Box Prediction Algorithms with Causal Learning
• Distributional Invariances and Interventional Markov Equivalence for Mixed Graph Models
• Conspiracy to Commit: Information Pollution, Artificial Intelligence, and Real-World Hate Crime
• Minimal enumeration of all possible total effects in a Markov equivalence class