Skip to main content

Gaussoids are two-antecedental approximations of Gaussian conditional independence structures

Abstract

The gaussoid axioms are conditional independence inference rules which characterize regular Gaussian CI structures over a three-element ground set. It is known that no finite set of inference rules completely describes regular Gaussian CI as the ground set grows. In this article we show that the gaussoid axioms logically imply every inference rule of at most two antecedents which is valid for regular Gaussians over any ground set. The proof is accomplished by exhibiting for each inclusion-minimal gaussoid extension of at most two CI statements a regular Gaussian realization. Moreover we prove that all those gaussoids have rational positive-definite realizations inside every ε-ball around the identity matrix. For the proof we introduce the concept of algebraic Gaussians over arbitrary fields and of positive Gaussians over ordered fields and obtain the same two-antecedental completeness of the gaussoid axioms for algebraic and positive Gaussians over all fields of characteristic zero as a byproduct.

References

  1. 1.

    Abo Khamis, M., Kolaitis, P.G., Ngo, H.Q., Suciu, D.: Decision problems in information theory. arXiv:2004.08783 (2020)

  2. 2.

    Boege, T.: Incidencegeometryintheprojectiveplaneviaalmost-principalminorsofsymmetricmatrices. arXiv:2103.02589 (2021)

  3. 3.

    Boege, T., D’Alì, A., Kahle, T., Sturmfels, B.: Thegeometryofgaussoids. Found.Comput.Math. 19(4), 775–812 (2019). https://doi.org/10.1007/s10208-018-9396-x

    MathSciNet  Article  Google Scholar 

  4. 4.

    Boege, T., Kahle, T.: Constructionmethodsforgaussoids. Kybernetika 56(6), 1045–1062 (2020). https://doi.org/10.14736/kyb-2020-6-1045

    MathSciNet  Article  Google Scholar 

  5. 5.

    Bouckaert, R., Hemmecke, R., Lindner, S., Studený, M.: Efficientalgorithmsforconditionalindependenceinference. J.Mach.Learn.Res. 11, 3453–3479 (2010)

    MathSciNet  Google Scholar 

  6. 6.

    Cox, D.A., Little, J., O’Shea, D.: Ideals,varieties,andalgorithms,fourthedn.UndergraduateTextsinMathematics. Springer, Berlin (2015). https://doi.org/10.1007/978-3-319-16721-3.Anintroductiontocomputationalalgebraicgeometryandcommutativealgebra

    Book  Google Scholar 

  7. 7.

    Dawid, A.P.: Conditionalindependenceinstatisticaltheory. J.Roy.Statist.Soc.Ser.B 41(1), 1–31 (1979)

    MathSciNet  Google Scholar 

  8. 8.

    Dougherty, R., Freiling, C., Zeger, K.: Linearrankinequalitiesonfiveormorevariables. arXiv:0910.0284(2010)

  9. 9.

    Drton, M., Xiao, H.: Smoothnessofgaussianconditionalindependencemodels. In: AlgebraicMethodsinStatisticsandProbabilityII,ContemporaryMathematics. https://doi.org/10.1090/conm/516/10173, vol. 516, pp 155–177. AmericanMathematicalSociety (2010)

  10. 10.

    Geiger, D., Pearl, J.: Logicalandalgorithmicpropertiesofconditionalindependenceandgraphicalmodels. Ann.Stat. 21 (4), 2001–2021 (1993). https://doi.org/10.1214/aos/1176349407

    Article  Google Scholar 

  11. 11.

    Geiger, D., Pearl, J.: Logicalandalgorithmicpropertiesofconditionalindependenceandgraphicalmodels. Ann.Statist. 21(4), 2001–2021 (1993). https://doi.org/10.1214/aos/1176349407

    MathSciNet  Article  Google Scholar 

  12. 12.

    Gong, Z., Aldeen, M., Elsner, L.: AnoteonageneralizedCramer’srule. LinearAlgebraAppl. 340(1), 253–254 (2002)

    Google Scholar 

  13. 13.

    Grayson, D.R., Stillman, M.E.: Macaulay2,asoftwaresystemforresearchinalgebraicgeometry.Availableathttp://www.math.uiuc.edu/Macaulay2/

  14. 14.

    Holtz, O., Sturmfels, B.: Hyperdeterminantalrelationsamongsymmetricprincipalminors.J.Algebra316(2). https://doi.org/10.1016/j.jalgebra.2007.01.039 (2007)

  15. 15.

    Lauritzen, S.: Graphicalmodels.,OxfordStatisticalScienceseries, vol. 17. OxfordUniversityPress, Oxford (1996)

    Google Scholar 

  16. 16.

    Lauritzen, S., Sadeghi, K.: UnifyingMarkovpropertiesforgraphicalmodels. Ann.Statist. 46 (5), 2251–2278 (2018). https://doi.org/10.1214/17-AOS1618

    MathSciNet  Article  Google Scholar 

  17. 17.

    Lněnička, R., Matúš, F.: OnGaussianconditionalindependencestructures. Kybernetika 43(3), 327–342 (2007)

    MathSciNet  Google Scholar 

  18. 18.

    Marshall, M.: Positivepolynomialsandsumsofsquares,MathematicalSurveysandmonographs, vol.146AmericanMathematicalSociety (2008)

  19. 19.

    Matúš, F.: Probabilisticconditionalindependencestructuresandmatroidtheory:background. Int.J.Gen.Syst. 22, 185–196 (1994)

    Article  Google Scholar 

  20. 20.

    Matúš, F.: Conditionalindependencestructuresexaminedviaminors. Ann.Math.Artif.Intell. 21(1), 99–30 (1997). https://doi.org/10.1023/A:1018957117081

    MathSciNet  Article  Google Scholar 

  21. 21.

    Matúš, F.: Towardsclassificationofsemigraphoids. DiscreteMath. 277(1), 115–145 (2004). https://doi.org/10.1016/S0012-365X(03)00155-9

    Article  Google Scholar 

  22. 22.

    Matúš, F.: Conditionalindependencesingaussianvectorsandringsofpolynomials. In: Kern-Isberner, G., Rödder, W., Kulmann, F. (eds.) Conditionals,Information,andInference,pp.152–161.Springer (2005)

  23. 23.

    Matúš, F.: Conditionalindependencesamongfourrandomvariables.II. Combin.Probab.Comput. 4(4), 407–417 (1995). https://doi.org/10.1017/S0963548300001747

    MathSciNet  Article  Google Scholar 

  24. 24.

    Matúš, F.: Conditionalindependencesamongfourrandomvariables.III.Finalconclusion. Combin.Probab.Comput. 8(3), 269–276 (1999). https://doi.org/10.1017/S0963548399003740

    MathSciNet  Article  Google Scholar 

  25. 25.

    Matúš, F., Studený, M.: Conditionalindependencesamongfourrandomvariables. I.Combin.Probab.Comput. 4(3), 269–278 (1995). https://doi.org/10.1017/S0963548300001644

    Article  Google Scholar 

  26. 26.

    Niepert, M., Gyssens, M., Sayrafi, B., VanGucht, D.: Ontheconditionalindependenceimplicationproblem:alattice-theoreticapproach. Artif.Intell. 202, 29–51 (2013). https://doi.org/10.1016/j.artint.2013.06.005

    Article  Google Scholar 

  27. 27.

    Oxley, J.: Matroidtheory,OxfordGraduateTextsinMathematics, 2nd edn., vol. 21. OxfordUniversityPress, Oxford (2011). https://doi.org/10.1093/acprof:oso/9780198566946.001.0001

  28. 28.

    Pearl, J.: Probabilisticreasoninginintelligentsystems:networksofplausibleinference. TheMorganKaufmannSeriesinRepresentationandReasoning.MorganKaufmann,SanMateoCA (1988)

  29. 29.

    Pearl, J., Paz, A.: GRAPHOIDS:AGraph-basedlogicforreasoningaboutrelevancerelations,orWhenwouldxtellyoumoreaboutyifyoualreadyknowz.Tech.Rep.CSD-850038UCLAComputerScience Department (1985)

  30. 30.

    Šimeček, P.: ClassesofGaussian,discreteandbinaryrepresentableindependencemodelshavenofinitecharacterization. In: ProceedingsofPragueStochastics,vol.400,pp.622–631 (2006)

  31. 31.

    Šimeček, P.: Gaussianrepresentationofindependencemodelsoverfourrandomvariables. In: COMPSTATConference (2006)

  32. 32.

    Šimeček, P.: Ashortnoteondiscreterepresentabilityofindependencemodels. In: ProccedingsoftheEuropeanWorkshoponProbabilisticGraphicalModels,pp.287–292 (2006)

  33. 33.

    Studený, M.: Probabilisticconditionalindependencestructures.Informationscienceandstatisticsspringer (2005)

  34. 34.

    Studený, M.: Conditionalindependenceandbasicmarkovproperties. In: Maathuis, M., Drton, M., Lauritzen, S., Wainwright, M. (eds.) Handbookofgraphicalmodels,Chapman&Hall/CRCHandbooksofModernStatisticalMethods,pp.3–38.CRCPress (2019)

  35. 35.

    Studený, M.: Conditionalindependencerelationshavenofinitecompletecharacterization.In: InformationTheory,StatisticalDecisionFunctionsandRandomProcesses,Vol.B,pp.377–396.Kluwer (1992)

  36. 36.

    Studený, M.: Semigraphoidsaretwo-antecedentalapproximationsofstochasticconditionalindependencemodels. In: DeMantaras, R.L., Poole, D., Proceedings, Uncertainty (eds.) , pp 546–552, MorganKaufmann (1994)

  37. 37.

    Sullivant, S.: Gaussianconditionalindependencerelationshavenofinitecompletecharacterization. J.PureAppl.Algebra 213(8), 1502–1506 (2009)

    Article  Google Scholar 

  38. 38.

    Sullivant, S.: Algebraicstatistics,GraduateStudiesinMathematics,vol.194AmericanMathematicalSociety (2018)

  39. 39.

    Verma, T., Pearl, J.: Analgorithmfordecidingifasetofobservedindependencieshasacausalexplanation. In: Dubois, D., Wellman, M.P., D’Ambrosio, B., Smets, P. (eds.) UncertaintyinArtificialIntelligence. https://doi.org/10.1016/B978-1-4832-8287-9.50049-9, pp 323–330, MorganKaufmann (1992)

Download references

Acknowledgements

The author wishes to thank Thomas Kahle, Andreas Kretschmer and Milan Studený for helpful discussions, and the anonymous referees for their clear and thoughtful suggestions for improvement of the earlier manuscript. Thanks to Xiangying Chen for spotting an error in the previous version of Example 7.4.

Funding

Open Access funding enabled and organized by Projekt DEAL. This work is funded by the Deutsche Forschungsgemeinschaft (DFG, German Research Foundation) – 314838170, GRK 2297 MathCoRe.

Author information

Affiliations

Authors

Corresponding author

Correspondence to Tobias Boege.

Ethics declarations

Conflict of Interests

The author declares that he has no conflict of interest.

Additional information

Publisher’snote

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Rights and permissions

Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/.

Reprints and Permissions

About this article

Verify currency and authenticity via CrossMark

Cite this article

Boege, T. Gaussoids are two-antecedental approximations of Gaussian conditional independence structures. Ann Math Artif Intell (2021). https://doi.org/10.1007/s10472-021-09780-0

Download citation

  • Accepted:

  • Published:

  • DOI: https://doi.org/10.1007/s10472-021-09780-0

Keywords

  • Conditional independence
  • Inference
  • Completeness
  • Gaussoid
  • Realizability
  • Rationality

Mathematics Subject Classification (2010)

  • 62B10
  • 62R01
  • 14P10