Australasian Joint Conference on Artificial Intelligence

AI 2015: Advances in Artificial Intelligence pp 411-423 | Cite as

Exploiting Innocuousness in Bayesian Networks

Conference paper
Part of the Lecture Notes in Computer Science book series (LNCS, volume 9457)

Abstract

Boolean combination functions in Bayesian networks, such as noisy-or, are often credited a property stating that inactive dependences (e.g., observed to false) do not “cause any harm” and an arc becomes vacuous and could have been left out. However, in classic Bayesian networks we are not able to express this property in local CPDs. By using novel ADBNs, we formalize the innocuousness property in CPDs and extend previous work on context-specific independencies. With an explicit representation of innocuousness in local CPDs, we provide a higher causal accuracy for CPD specifications and open new ways for more efficient and less-restricted reasoning in (A)DBNs.

References

  1. 1.
    Antonucci, A.: The imprecise noisy-OR gate. In: 14th International Conference on Information Fusion, pp. 1–7. IEEE (2011)Google Scholar
  2. 2.
    Boutilier, C., Friedman, N., Goldszmidt, M., Koller, D.: Context-specific independence in Bayesian networks. In: 12th Conference on Uncertainty in Artificial Intelligence, pp. 115–123 (1996)Google Scholar
  3. 3.
    Cozman, F.G.: Axiomatizing noisy-OR. In: 16th Eureopean Conference on Artificial Intelligence, p. 979 (2004)Google Scholar
  4. 4.
    Heckerman, D., Breese, J.S.: Causal independence for probability assessment and inference using Bayesian networks. IEEE Trans. Syst. Man Cybern. Part A Syst. Hum. 26(6), 826–831 (1996)CrossRefGoogle Scholar
  5. 5.
    Henrion, M.: Practical issues in constructing a Bayes belief network. Int. J. Approximate Reasoning 2(3), 337 (1988)MathSciNetGoogle Scholar
  6. 6.
    Jaeger, M.: Relational Bayesian networks. In: 13th Conference on Uncertainty in Artificial Intelligence, pp. 266–273 (1997)Google Scholar
  7. 7.
    Motzek, A., Möller, R.: Indirect causes in dynamic Bayesian networks revisited. In: 24th International Joint Conference on Artificial Intelligence, pp. 703–709. AAAI (2015)Google Scholar
  8. 8.
    Pearl, J.: Reasoning with cause and effect. AI Mag. 23(1), 1–83 (2002)Google Scholar
  9. 9.
    Poole, D., Zhang, N.L.: Exploiting contextual independence in probabilistic inference. J. Artif. Intell. Res. 18, 263–313 (2003)MathSciNetMATHGoogle Scholar
  10. 10.
    Sloane, N.J.A.: The on-line encyclopedia of integer sequences. OEIS Foundation Inc., Sequences A003024 & A001831 (2015). http://oeis.org/
  11. 11.
    Srinivas, S.: A generalization of the noisy-OR model. In: 9th International Conference on Uncertainty in Artificial Intelligence, pp. 208–215 (1993)Google Scholar
  12. 12.
    Zagorecki, A., Druzdzel, M.J.: Probabilistic independence of causal influences. In: 3rd European Workshop on Probabilistic Graphical Models, pp. 325–332 (2006)Google Scholar
  13. 13.
    Zhang, N.L., Poole, D.: Exploiting causal independence in Bayesian network inference. J. Artif. Intell. Res. 5, 301–328 (1996)MathSciNetMATHGoogle Scholar

Copyright information

© Springer International Publishing Switzerland 2015

Authors and Affiliations

  1. 1.Institute of Information SystemsUniversität zu LübeckLübeckGermany

Personalised recommendations