Corpus ID: 52050883

Causal Discovery by Telling Apart Parents and Children

@article{Marx2018CausalDB,
  title={Causal Discovery by Telling Apart Parents and Children},
  author={A. Marx and J. Vreeken},
  journal={ArXiv},
  year={2018},
  volume={abs/1808.06356}
}
  • A. Marx, J. Vreeken
  • Published 2018
  • Mathematics, Computer Science
  • ArXiv
  • We consider the problem of inferring the directed, causal graph from observational data, assuming no hidden confounders. We take an information theoretic approach, and make three main contributions. First, we show how through algorithmic information theory we can obtain SCI, a highly robust, effective and computationally efficient test for conditional independence---and show it outperforms the state of the art when applied in constraint-based inference methods such as stable PC. Second… CONTINUE READING
    Testing Conditional Independence on Discrete Data using Stochastic Complexity
    • 4
    • PDF

    References

    Publications referenced by this paper.
    SHOWING 1-10 OF 31 REFERENCES
    Probabilistic reasoning in intelligent systems - networks of plausible inference
    • J. Pearl
    • Computer Science, Philosophy
    • 1989
    • 17,402
    • Highly Influential
    • PDF
    Elements of Information Theory
    • 38,850
    • PDF
    Optimal Structure Identification With Greedy Search
    • 1,057
    • Highly Influential
    • PDF
    Local Causal and Markov Blanket Induction for Causal Discovery and Feature Selection for Classification Part I: Algorithms and Empirical Evaluation
    • 383
    • Highly Influential
    • PDF
    An Introduction to Kolmogorov Complexity and Its Applications
    • 3,531
    • PDF
    Causal Inference Using the Algorithmic Markov Condition
    • 155
    • Highly Influential
    • PDF
    Causation, Prediction, and Search
    • T. Burr
    • Computer Science, Mathematics
    • 2003
    • 1,740
    • PDF
    Bayesian Network Induction via Local Neighborhoods
    • 280
    • PDF
    Order-independent constraint-based causal structure learning
    • 232
    • Highly Influential
    • PDF
    Computation, causation, and discovery
    • 326