BaCaDI: Bayesian Causal Discovery with Unknown Interventions

2022/06/03 by Alexander Hägele, Hägele, Alexander, Jonas Rothfuss +9 · 8 citations
Computer Science · #Bayesian Modeling and Causal Inference #FOS: Computer and information sciences #Machine Learning (cs.LG) #Machine Learning (stat.ML) #Machine Learning and Algorithms #Machine Learning and Data Classification #Methodology (stat.ME)

paper · pdf · doi:10.48550/arxiv.2206.01665

openalex publication_date 2022/06/03 · openalex created_date 2025/10/10 · openalex updated_date 2026/07/28

Abstract

Inferring causal structures from experimentation is a central task in many domains. For example, in biology, recent advances allow us to obtain single-cell expression data under multiple interventions such as drugs or gene knockouts. However, the targets of the interventions are often uncertain or unknown and the number of observations limited. As a result, standard causal discovery methods can no longer be reliably used. To fill this gap, we propose a Bayesian framework (BaCaDI) for discovering and reasoning about the causal structure that underlies data generated under various unknown experimental or interventional conditions. BaCaDI is fully differentiable, which allows us to infer the complex joint posterior over the intervention targets and the causal structure via efficient gradient-based variational inference. In experiments on synthetic causal discovery tasks and simulated gene-expression data, BaCaDI outperforms related methods in identifying causal structures and intervention targets.

Cited by

• PerturbPFN: Probing the Limits of Synthetic Priors in Drug Perturbation Modelling
• Amortized Bayesian Causal Discovery of Extended Factor Graphs
• MetaCaDI: A Meta-Learning Framework for Scalable Causal Discovery with Unknown Interventions
• Causal-Guided Dimension Reduction for Efficient Pareto Optimization
• Large-Scale Bayesian Causal Discovery with Interventional Data
• Causal Effect Identification in Heterogeneous Environments from Higher-Order Moments
• Bayesian Hierarchical Invariant Prediction
• Towards Interpretable Deep Generative Models via Causal Representation Learning

Related

• Interventions, Where and How? Experimental Design for Causal Models at Scale
• MetaCaDI: A Meta-Learning Framework for Scalable Causal Discovery with Unknown Interventions
• Bayesian causal discovery from unknown general interventions
• Bayesian Intervention Optimization for Causal Discovery
• Large-Scale Bayesian Causal Discovery with Interventional Data