Skip to main content
Log in

Model Failure and Context Switching Using Logic-Based Stochastic Models

  • Regular Paper
  • Published:
Journal of Computer Science and Technology Aims and scope Submit manuscript

Abstract

This paper addresses parameter drift in stochastic models. We define a notion of context that represents invariant, stable-over-time behavior and we then propose an algorithm for detecting context changes in processing a stream of data. A context change is seen as model failure, when a probabilistic model representing current behavior is no longer able to “fit” newly encountered data. We specify our stochastic models using a first-order logic-based probabilistic modeling language called Generalized Loopy Logic (GLL). An important component of GLL is its learning mechanism that can identify context drift. We demonstrate how our algorithm can be incorporated into a failure-driven context-switching probabilistic modeling framework and offer several examples of its application.

This is a preview of subscription content, log in via an institution to check access.

Access this article

Price excludes VAT (USA)
Tax calculation will be finalised during checkout.

Instant access to the full article PDF.

Institutional subscriptions

Similar content being viewed by others

References

  1. Wellman M P, Breese J S, Goldman R P. From knowledge bases to decision models. Knowledge Engineering Review, 1992, 7(1): 35-53.

    Article  Google Scholar 

  2. Piaget J. Piaget’s Theory. Handbook of Child Psychology, Mussen P (ed.), 4th Edition, New York: Wiley, 1983.

  3. Poole D. Logic programming, abduction and probability: A top-down anytime algorithm for estimating prior and posterior probabilities. New Generation Computing, 1993, 11(3/4): 377-400.

    Article  MATH  Google Scholar 

  4. Haddawy P. Generating Bayesian networks from probability logic knowledge bases. In Proc. the 10th Conf. Uncertainty in AI, Seattle, USA, Jul. 29-31, 1994, pp.262-269.

  5. Ngo L, Haddawy P. Answering queries from context-sensitive probabilistic knowledge bases. Theoretical Computer Science, 1997, 171(1/2): 147-177.

    Article  MathSciNet  MATH  Google Scholar 

  6. Ngo L, Haddawy P, Krieger R A, Helwig J. Efficient temporal probabilistic reasoning via context-sensitive model construction. Computers in Biology and Medicine, 1997, 27(5): 453-476.

    Article  Google Scholar 

  7. Glesner S, Koller D. Constructing flexible dynamic belief networks from first-order probabilistic knowledge bases. In Proc. the European Conference on Symbolic and Quantitative Approaches to Reasoning and Uncertainty, Fribourg, Switzerland, July 3-5, 1995, pp.217-226.

  8. Sato T, Kameya Y. PRISM: A language for symbolicstatistical modeling. In Proc. 15th Int. Joint Conf. AI (IJCAI), Nagoya, Japan, Aug. 23-29, 1997, pp.1330-1335.

  9. Sato T, Kameya Y. New advances in logic-based probabilistic modeling by PRISM. Probabilistic Inductive Logic Programming, LNCS 4911, Springer, 2008, pp.118-155.

  10. DeRaedt L, Kimmig A, Toivonen H. ProbLog: A probabilistic Prolog and its application in link discovery. In Proc. the 20th Int. Joint Conf. AI (IJCAI), Hyderabad, India, Jan. 6-12, 2007, pp.2468-2473.

  11. Friedman N, Getoor L, Koller D, Pfeffer A. Learning probabilistic relational models. In Proc. 16th Int. Joint Conf. AI (IJCAI), Stockholm, Sweden, Jul. 31-Aug. 6, 1999, pp.1300-1307.

  12. Getoor L, Friedman N, Koller D, Pfeffer A. Learning Probabilistic Relational Models. Relational Data Mining, 2001, Springer-Verlag, New York, pp.307-335.

  13. Kersting K, DeRaedt L. Bayesian logic programs. In Proc. the 10th Int. Conf. ILP, London, UK, July 24-27, 2000, pp.138-155.

  14. Richardson M, Domingos P. Markov logic networks. Machine Learning, 2006, 62(1/2): 107-136.

    Article  Google Scholar 

  15. Shen Y D. Reasoning with recursive loops under the PLP framework. ACM Trans. Computational Logic, 2008, 9(4): 1-30.

    Article  Google Scholar 

  16. Sanghai S, Domingos P,Weld D. Relational dynamic Bayesian networks. Artificial Intelligence Research, 2005, 24: 759-797.

    MATH  Google Scholar 

  17. Turney P. The identification of context-sensitive features: A formal definition of context for concept learning. In Proc. Workshop on Learning in Context-Sensitive Domains at the 13th ICML, Bari, Italy, Jul. 3-6, 1996, pp.53-59.

  18. Sanscartier M J, Neufeld E. Identifying hidden variables from context-specific independencies. In Proc. FLAIRS 2007, Key West, USA, May 7-9, 2007, pp.472-477.

  19. Pearl J. Causality: Models, Reasoning, and Inference. Cambridge University Press, 2000.

  20. Halpern J, Pearl J. Causes and explanations: A structuralmodel approach — Part 1: Causes. In Proc. the 17th Conf. Uncertainty in AI (UAI 2001), Seattle, USA, Aug. 2-5, 2001, pp.194-202.

  21. Pearl J. Probabilistic Reasoning in Intelligent Systems: Networks of Plausible Inference. Morgan Kaufmann, 1988.

  22. Chickering D. Learning Bayesian networks is NP-complete. In Proc. AI and Stat., Fort Lauderdale, USA, Jan. 1995, pp.121-130.

    Google Scholar 

  23. Pless D J, Chakrabarti C, Rammohan R, Luger G F. The design and testing of a first-order stochastic modeling language. International Journal on Artificial Intelligence Tools, 2006, 15(6): 979-1005.

    Article  Google Scholar 

  24. Dempster A, Laird N, Rubin D. Maximum likelihood from incomplete data via the EM algorithm. J. the Royal Statistical Society, Series B (Methodological), 1977, 39(1): 1-38.

    MathSciNet  MATH  Google Scholar 

  25. Luger G F. Artificial Intelligence: Structures and Strategies for Complex Problem Solving. Addison-Wesley, 2009.

  26. Murphy K P, Weiss Y, Jordan M. Loopy Belief Propagation for Approximate Inference: An Empirical Study. In Proc. Uncertainty in Artificial Intelligence, Stockholm, Sweden, July 30-Aug. 1, 1999, pp.467-475.

  27. Pollak M. Optimal detection of a change in distribution. The Annals of Statistics, 1985, 13(1): 206-227.

    Article  MathSciNet  MATH  Google Scholar 

  28. Dayanik S, Goulding C, Poor H V. Joint detection and identification of an unobservable change in the distribution of a random sequence. In Proc. Conf. Information Sciences and Systems, Baltimore, USA, Mar. 14-16, 2007, pp.68-73.

  29. Steyvers M, Brown S. Prediction and change detection. Advances in Neural Information Processing Systems, 2006, 18: 1281-1288.

    Google Scholar 

  30. Song X, Wu M, Jermaine C, Ranka S. Statistical change detection for multi-dimensional data. In Proc. the 13th Int. Conf. Knowledge Discovery and Data Mining (KDD2007), 2007, pp.667-676.

  31. Kuncheva L I. Classifier ensembles for detecting concept change in streaming data: Overview and perspectives. In Proc. the 2nd Workshop SUEMA (ECAI), Patras, Greece, Jul. 21-25, 2008, pp.5-10.

  32. Dybvig R K. The Scheme Programming Language. The MIT Press, 2009.

Download references

Author information

Authors and Affiliations

Authors

Corresponding author

Correspondence to Nikita A. Sakhanenko.

Additional information

This work was funded by a US Air Force Research Laboratory SBIR contract (FA8750-06-C0016).

Rights and permissions

Reprints and permissions

About this article

Cite this article

Sakhanenko, N.A., Luger, G.F. Model Failure and Context Switching Using Logic-Based Stochastic Models. J. Comput. Sci. Technol. 25, 665–680 (2010). https://doi.org/10.1007/s11390-010-9356-7

Download citation

  • Received:

  • Revised:

  • Published:

  • Issue Date:

  • DOI: https://doi.org/10.1007/s11390-010-9356-7

Keywords

Navigation