mlsanityData to 14 Oct 2026

SVI-DAG: A Structured Variational Inference Approach to Bayesian Causal Discovery

arXiv 2608.04930 ↗First posted 5 Aug 2026cs.LG cs.AIcited by 0

Not on the map. Why?

Abstract (arXiv)

Bayesian causal discovery seeks to determine the posterior distribution of causal theories, which are interpreted as directed acyclic graphs (DAGs) that explain the observed data. The resulting posterior allows systematic reasoning regarding epistemic uncertainty within these theories. Nonetheless, finding such graphs is difficult due to identifiability problems and limited observational data. Furthermore, precisely approximating posterior over graphs is challenging given vast range of potential DAGs. Recent Bayesian approaches have addressed some of these challenges, yet they remain limited as they fail to encode dependencies between edges, and lack principled ways to incorporate domain knowledge as inductive biases during the search process. To overcome these limitations, we propose SVI-DAG, a structured variational inference approach to Bayesian causal discovery using observational data and prior beliefs that uses normalizing flows to model dependencies between edges, supporting expressive and multimodal posterior learning over DAGs. To mitigate mode seeking behaviour in evidence lower bound optimization and promote mode coverage, we use stein variational gradient descent to update the node potentials using a kernel in acyclicity space. We evaluate SVI-DAG against 5 state-of-the-art Bayesian DAG learning methods and demonstrate competitive performance in terms of both accuracy and uncertainty quantification.

Key facts. No citing papers in the map yet (data through 14 Oct 2026). Builds on 38 of its 52 references that are in the map. First posted on arXiv 5 Aug 2026.

Builds on

38 of 52 references.

References · 38

  1. 2016Categorical Reparameterization with Gumbel-Softmaxcited by 2,739
  2. 2015Variational Inference with Normalizing Flowscited by 2,170
  3. 2016Variational Inference: A Review for Statisticianscited by 1,472
  4. 2016The Concrete Distribution: A Continuous Relaxation of Discrete Random Variablescited by 1,148
  5. 2018DAGs with NO TEARS: Continuous Optimization for Structure Learningcited by 752
  6. 2019Neural Spline Flowscited by 596
  7. 2016Stein Variational Gradient Descent: A General Purpose Bayesian Inference Algorithmcited by 507
  8. 2019DAG-GNN: DAG Structure Learning with Graph Neural Networkscited by 331
  9. 2013Causal Discovery with Continuous Additive Noise Modelscited by 255
  10. 2017Advances in Variational Inferencecited by 236
  11. 2019Gradient-Based Neural DAG Learningcited by 227
  12. 2021D'ya like DAGs? A Survey on Structure Learning and Causal Discoverycited by 211
26 more
  1. 2020On the Role of Sparsity and DAG Constraints for Learning Linear DAGscited by 191
  2. 2022Bayesian Structure Learning with Generative Flow Networkscited by 153
  3. 2022DAGMA: Learning DAGs via M-matrices and a Log-Determinant Acyclicity Characterizationcited by 140
  4. 2021DiBS: Differentiable Bayesian Structure Learningcited by 106
  5. 2020Amortized Causal Discovery: Learning to Infer Causal Graphs from Time-Series Datacited by 100
  6. 2019D-VAE: A Variational Autoencoder for Directed Acyclic Graphscited by 98
  7. 2022Amortized Inference for Causal Structure Learningcited by 90
  8. 2019Masked Gradient-Based Causal Structure Learningcited by 89
  9. 2017Causal Structure Learningcited by 84
  10. 2025Superintelligent Agents Pose Catastrophic Risks: Can Scientist AI Offer a Safer Path?cited by 75
  11. 2021BCD Nets: Scalable Variational Approaches for Bayesian Causal Discoverycited by 68
  12. 2019A Graph Autoencoder Approach to Causal Structure Learningcited by 55
  13. 2021DAGs with No Curl: An Efficient DAG Structure Learning Approachcited by 53
  14. 2023Joint Bayesian Inference of Graphical Structure and Parameters with a Single Generative Flow Networkcited by 50
  15. 2022Differentiable DAG Samplingcited by 47
  16. 2022Causal Structure Learning: a Combinatorial Perspectivecited by 46
  17. 2021Variational Causal Networks: Approximate Bayesian Inference over Causal Structurescited by 43
  18. 2023Fast Scalable and Accurate Discovery of DAGs Using the Best Order Score Search and Grow-Shrink Treescited by 38
  19. 2023BayesDAG: Gradient-Based Posterior Inference for Causal Discoverycited by 33
  20. 2022Bayesian learning of Causal Structure and Mechanisms with GFlowNets and Variational Bayescited by 30
  21. 2024A Meta-Learning Approach to Bayesian Causal Discoverycited by 22
  22. 2026International AI Safety Report 2026cited by 22
  23. 2023NODAGS-Flow: Nonlinear Cyclic Causal Structure Learningcited by 20
  24. 2022Tractable Uncertainty for Structure Learningcited by 15
  25. 2024ProDAG: Projected Variational Inference for Directed Acyclic Graphscited by 4
  26. 2025Differentiable Constraint-Based Causal Discoverycited by 4

Built on by

Not cited yet.