Differentiable Causal Discovery Under Unmeasured Confounding

Abstract

The data drawn from biological, economic, and social systems are often confounded due to the presence of unmeasured variables. Prior work in causal discovery has focused on discrete search procedures for selecting acyclic directed mixed graphs (ADMGs), specifically ancestral ADMGs, that encode ordinary conditional independence constraints among the observed variables of the system. However, confounded systems also exhibit more general equality restrictions that cannot be represented via these graphs, placing a limit on the kinds of structures that can be learned using ancestral ADMGs. In this work, we derive differentiable algebraic constraints that fully characterize the space of ancestral ADMGs, as well as more general classes of ADMGs, arid ADMGs and bow-free ADMGs, that capture all equality restrictions on the observed variables. We use these constraints to cast causal discovery as a continuous optimization problem and design differentiable procedures to find the best fitting ADMG when the data comes from a confounded linear system of equations with correlated errors. We demonstrate the efficacy of our method through simulations and application to a protein expression dataset. Code implementing our methods is open-source and publicly available at https://gitlab.com/rbhatta8/dcd and will be incorporated into the Ananke package.

Main draft: 9 pages. Appendix: 9 pages

References in corpus (13)

• DAGs with NO TEARS: Continuous Optimization for Structure Learning
• DAG-GNN: DAG Structure Learning with Graph Neural Networks
• Pearl's Calculus of Intervention Is Complete
• Global identifiability of linear structural equation models
• Nested Markov Properties for Acyclic Directed Mixed Graphs
• Markovian acyclic directed mixed graphs for discrete data
• Margins of discrete Bayesian networks
• Computing maximum likelihood estimates in recursive linear models with correlated errors
• Model selection and local geometry
• Semiparametric Inference For Causal Effects In Graphical Models With Hidden Variables
• FASK with Interventional Knowledge Recovers Edges from the Sachs Model
• Causal Discovery with Unobserved Confounding and non-Gaussian Data
• Characterizing Distribution Equivalence and Structure Learning for Cyclic and Acyclic Directed Graphs

Cited by in corpus (10)

• Nested Markov Properties for Acyclic Directed Mixed Graphs
• gCastle: A Python Toolbox for Causal Discovery
• Causal Inference in the Presence of Interference in Sponsored Search Advertising
• Beware of the Simulated DAG! Causal Discovery Benchmarks May Be Easy To Game
• Towards Federated Bayesian Network Structure Learning with Continuous Optimization
• Unsuitability of NOTEARS for Causal Graph Discovery
• CausalCity: Complex Simulations with Agency for Causal Discovery and Reasoning
• Integer Programming for Causal Structure Learning in the Presence of Latent Variables
• On the Convergence of Continuous Constrained Optimization for Structure Learning
• Entropic Inequality Constraints from $e$-separation Relations in Directed Acyclic Graphs with Hidden Variables