mlsanityData to 7 Oct 2026

The impact of prior knowledge on causal structure learning

arXiv 2102.00473 ↗First posted 31 Jan 2021cs.AI cs.LGcited by 35

Not on the map. Why?

Abstract (arXiv)

Causal Bayesian networks have become a powerful technology for reasoning under uncertainty in areas that require transparency and explainability, by relying on causal assumptions that enable us to simulate hypothetical interventions. The graphical structure of such models can be estimated by structure learning algorithms, domain knowledge, or a combination of both. Various knowledge approaches have been proposed in the literature that enable us to specify prior knowledge that constrains or guides these algorithms. This paper introduces some novel, and also describes some existing, knowledge-based approaches that enable us to combine structure learning with knowledge obtained from heterogeneous sources. We investigate the impact of these approaches on structure learning across different algorithms, case studies and settings that we might encounter in practice. Each approach is assessed in terms of effectiveness and efficiency, including graphical accuracy, model fitting, complexity, and runtime; making this the first paper that provides a comparative evaluation of a wide range of knowledge approaches for structure learning. Because the value of knowledge depends on what data are available, we illustrate the results both with limited and big data. While the overall results show that knowledge becomes less important with big data due to higher learning accuracy rendering knowledge less important, some of the knowledge approaches are found to be more important with big data. Amongst the main conclusions is the observation that reduced search space obtained from knowledge does not always imply reduced computational complexity, perhaps because the relationships implied by the data and knowledge are in tension.

Key facts. Cited by 35 papers in the map (data through 7 Oct 2026). Builds on 9 of its 61 references that are in the map. First posted on arXiv 31 Jan 2021.

Builds on

9 of 61 references.

References · 9

  1. 2013Learning Bayesian Networks: The Combination of Knowledge and Statistical Datacited by 185
  2. 2020Large-scale empirical validation of Bayesian Network structure learning algorithms with noisy datacited by 29
  3. 2013Learning Bayesian Networks: A Unification for Discrete and Gaussian Domainscited by 21
  4. 2013Properties of Bayesian Belief Network Learning Algorithmscited by 13
  5. 2016Formalizing Neurath's Ship: Approximate Algorithms for Online Causal Learningcited by 12
  6. 2016An Empirical-Bayes Score for Discrete Bayesian Networkscited by 9
  7. 2020Learning Bayesian Networks that enable full propagation of evidencecited by 7
  8. 2019Learning Bayesian networks from demographic and health survey datacited by 6
  9. 2020Improving Bayesian Network Structure Learning in the Presence of Measurement Errorcited by 5

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2021: 1; 2022: 3; 2023: 7; 2024: 6; 2025: 10; 2026: 8.

35 papers.

Sort:

By area:

  1. 2021A survey of Bayesian Network structure learningcited by 85
  2. 2023Causal Discovery with Language Models as Imperfect Expertscited by 62
  3. 2023Integrating Large Language Model for Improved Causal Discoverycited by 54
  4. 2023Causal Order: The Key to Leveraging Imperfect Experts in Causal Inferencecited by 46
  5. 2024ALCM: Autonomous LLM-Augmented Causal Discovery Frameworkcited by 29
  6. 2023Causal Structure Learning Supervised by Large Language Modelcited by 21
  7. 2024Large Language Models for Constrained-Based Causal Discoverycited by 15
  8. 2024CausalGraph2LLM: Evaluating LLMs for Causal Queriescited by 12
  9. 2024bnRep: A repository of Bayesian networks from the academic literaturecited by 9
  10. 2022Causal Discovery and Knowledge Injection for Contestable Neural Networks (with Appendices)cited by 6
  11. 2023Mitigating Prior Errors in Causal Structure Learning: A Resilient Approach via Bayesian Networkscited by 5
  12. 2023Causal discovery using dynamically requested knowledgecited by 5
23 more
  1. 2025Bayesian Network Structure Discovery Using Large Language Modelscited by 5
  2. 2022Domain Knowledge in A*-Based Causal Discoverycited by 4
  3. 2024Causal Discovery over High-Dimensional Structured Hypothesis Spaces with Causal Graph Partitioningcited by 4
  4. 2024Using GPT-4 to guide causal machine learningcited by 3
  5. 2026EDGE: Error Dependency Graph-Guided Multi-Error Attribution in Multi-Agent LLM Systemscited by 3
  6. 2022CKH: Causal Knowledge Hierarchy for Estimating Structural Causal Models from Data and Priorscited by 2
  7. 2025Learning to Defer for Causal Discovery with Imperfect Expertscited by 2
  8. 2025Robust Causal Discovery under Imperfect Structural Constraintscited by 2
  9. 2025Causal Discovery on Higher-Order Interactionscited by 2
  10. 2026Dynamic Expert-Guided Model Averaging for Causal Discoverycited by 2
  11. 2025Integrating Probabilistic Trees and Causal Networks for Clinical and Epidemiological Datacited by 1
  12. 2025Less Greedy Equivalence Searchcited by 1
  13. 2025Incorporating Expert Knowledge into Bayesian Causal Discovery of Mixtures of Directed Acyclic Graphscited by 1
  14. 2023Shapley-PC: Constraint-based Causal Structure Learning with a Shapley Inspired Frameworkcited by 0
  15. 2025Sequential Causal Discovery with Noisy Language Model Priorscited by 0
  16. 2025Towards Privacy-Aware Bayesian Networks: A Credal Approachcited by 0
  17. 2025Linear Causal Discovery with Interventional Constraintscited by 0
  18. 2026Causal Discovery Should Embrace the Wisdom of the Crowdcited by 0
  19. 2026Causal Transfer in Medical Image Analysiscited by 0
  20. 2026PRCD-MAP: Learning How Much to Trust Imperfect Priors in Causal Discoverycited by 0
  21. 2026Guide, Not Bind: Why Defeasible Priors Fail in Augmented Lagrangian Causal Discoverycited by 0
  22. 2026LEBGen: An LLM-Enhanced Bayesian Network Framework for Few-Shot Travel Survey Data Generationcited by 0
  23. 2026xWhyL: Causal Interactive Learningcited by 0