ML Anthology

Reasoning About Saturated Conditional Independence Under Uncertainty: Axioms, Algorithms, and Levesque's Situations to the Rescue

Sebastian Link
AAAI 2013 pp. 612-618
doi:10.1609/AAAI.V27I1.8564 /aaai/2013/link2013aaai-reasoning/

Abstract

The implication problem of probabilistic conditional independencies is investigated in the presence of missing data. Here, graph separation axioms fail to hold for saturated conditional independencies, unlike the known idealized case with no missing data. Several axiomatic, algorithmic, and logical characterizations of the implication problem for saturated conditional independencies are established. In particular, equivalences are shown to the implication problem of a propositional fragment under Levesque's situations, and that of Lien's class of multivalued database dependencies under null values.

PDF AAAI Semantic Scholar

Cite

Text

Link. "Reasoning About Saturated Conditional Independence Under Uncertainty: Axioms, Algorithms, and Levesque's Situations to the Rescue." AAAI Conference on Artificial Intelligence, 2013. doi:10.1609/AAAI.V27I1.8564

Markdown

[Link. "Reasoning About Saturated Conditional Independence Under Uncertainty: Axioms, Algorithms, and Levesque's Situations to the Rescue." AAAI Conference on Artificial Intelligence, 2013.](https://mlanthology.org/aaai/2013/link2013aaai-reasoning/) doi:10.1609/AAAI.V27I1.8564

BibTeX

@inproceedings{link2013aaai-reasoning,
  title     = {{Reasoning About Saturated Conditional Independence Under Uncertainty: Axioms, Algorithms, and Levesque's Situations to the Rescue}},
  author    = {Link, Sebastian},
  booktitle = {AAAI Conference on Artificial Intelligence},
  year      = {2013},
  pages     = {612-618},
  doi       = {10.1609/AAAI.V27I1.8564},
  url       = {https://mlanthology.org/aaai/2013/link2013aaai-reasoning/}
}