Discriminative Learning of First-Order Weighted Abduction from Partial Discourse Explanations

  • Kazeto Yamamoto
  • Naoya Inoue
  • Yotaro Watanabe
  • Naoaki Okazaki
  • Kentaro Inui
Part of the Lecture Notes in Computer Science book series (LNCS, volume 7816)

Abstract

Abduction is inference to the best explanation. Abduction has long been studied in a wide range of contexts and is widely used for modeling artificial intelligence systems, such as diagnostic systems and plan recognition systems. Recent advances in the techniques of automatic world knowledge acquisition and inference technique warrant applying abduction with large knowledge bases to real-life problems. However, less attention has been paid to how to automatically learn score functions, which rank candidate explanations in order of their plausibility. In this paper, we propose a novel approach for learning the score function of first-order logic-based weighted abduction [1] in a supervised manner. Because the manual annotation of abductive explanations (i.e. a set of literals that explains observations) is a time-consuming task in many cases, we propose a framework to learn the score function from partially annotated abductive explanations (i.e. a subset of those literals). More specifically, we assume that we apply abduction to a specific task, where a subset of the best explanation is associated with output labels, and the rest are regarded as hidden variables. We then formulate the learning problem as a task of discriminative structured learning with hidden variables. Our experiments show that our framework successfully reduces the loss in each iteration on a plan recognition dataset.

Preview

Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.

References

  1. 1.
    Hobbs, J.R., Stickel, M., Martin, P., Edwards, D.: Interpretation as abduction. Artificial Intelligence 63, 69–142 (1993)CrossRefGoogle Scholar
  2. 2.
    Fellbaum, C. (ed.): WordNet: an electronic lexical database. MIT Press (1998)Google Scholar
  3. 3.
    Ruppenhofer, J., Ellsworth, M., Petruck, M., Johnson, C., Scheffczyk, J.: FrameNet II: Extended Theory and Practice. Technical report, Berkeley, USA (2010)Google Scholar
  4. 4.
    Chambers, N., Jurafsky, D.: Unsupervised Learning of Narrative Schemas and their Participants. In: ACL, pp. 602–610 (2009)Google Scholar
  5. 5.
    Schoenmackers, S., Davis, J., Etzioni, O., Weld, D.: Learning First-order Horn Clauses from Web Text. In: EMNLP, pp. 1088–1098 (2010)Google Scholar
  6. 6.
    Hovy, D., Zhang, C., Hovy, E., Penas, A.: Unsupervised discovery of domain-specific knowledge from text. In: ACL, pp. 1466–1475 (2011)Google Scholar
  7. 7.
    Inoue, N., Inui, K.: ILP-Based Reasoning for Weighted Abduction. In: AAAI Workshop on Plan, Activity and Intent Recognition (2011)Google Scholar
  8. 8.
    Ovchinnikova, E., Montazeri, N., Alexandrov, T., Hobbs, J.R., McCord, M., Mulkar-Mehta, R.: Abductive Reasoning with a Large Knowledge Base for Discourse Processing. In: IWCS, Oxford, UK, pp. 225–234 (2011)Google Scholar
  9. 9.
    Dagan, I., Dolan, B., Magnini, B., Roth, D.: Recognizing textual entailment: Rational, evaluation and approaches - Erratum. NLE 16, 105 (2010)CrossRefGoogle Scholar
  10. 10.
    Kate, R.J., Mooney, R.J.: Probabilistic Abduction using Markov Logic Networks. In: PAIRS (2009)Google Scholar
  11. 11.
    Blythe, J., Hobbs, J.R., Domingos, P., Kate, R.J., Mooney, R.J.: Implementing Weighted Abduction in Markov Logic. In: IWCS, Oxford, UK, pp. 55–64 (2011)Google Scholar
  12. 12.
    Singla, P., Domingos, P.: Abductive Markov Logic for Plan Recognition. In: AAAI, pp. 1069–1075 (2011)Google Scholar
  13. 13.
    Richardson, M., Domingos, P.: Markov logic networks. In: ML, pp. 107–136 (2006)Google Scholar
  14. 14.
    Huynh, T.N., Mooney, R.J.: Max-Margin Weight Learning for Markov Logic Networks. In: Proceedings of the International Workshop on Statistical Relational Learning, SRL 2009 (2009)Google Scholar
  15. 15.
    Lowd, D., Domingos, P.: Efficient Weight Learning for Markov Logic Networks. In: Kok, J.N., Koronacki, J., Lopez de Mantaras, R., Matwin, S., Mladenič, D., Skowron, A. (eds.) PKDD 2007. LNCS (LNAI), vol. 4702, pp. 200–211. Springer, Heidelberg (2007)CrossRefGoogle Scholar
  16. 16.
    Charniak, E., Goldman, R.P.: A Probabilistic Model of Plan Recognition. In: AAAI, pp. 160–165 (1991)Google Scholar
  17. 17.
    Poole, D.: Probabilistic Horn abduction and Bayesian networks. Artificial Intelligence 64 (1), 81–129 (1993)MATHCrossRefGoogle Scholar
  18. 18.
    Raghavan, S., Mooney, R.J.: Bayesian Abductive Logic Programs. In: STARAI, pp. 82–87 (2010)Google Scholar
  19. 19.
    Ovchinnikova, E.: Integration of World Knowledge for Natural Language Understanding. Atlantis Press (2012)Google Scholar
  20. 20.
    Charniak, E., Shimony, S.E.: Probabilistic semantics for cost based abduction. In: AAAI, pp. 106–111 (1990)Google Scholar
  21. 21.
    Ng, H.T., Mooney, R.J.: Abductive Plan Recognition and Diagnosis: A Comprehensive Empirical Evaluation. In: KR, pp. 499–508 (1992)Google Scholar
  22. 22.
    Inoue, N., Inui, K.: Large-scale Cost-based Abduction in Full-fledged First-order Logic with Cutting Plane Inference. In: Proceedings of the 12th European Conference on Logics in Artificial Intelligence (2012) (to appear)Google Scholar
  23. 23.
    Poon, H., Domingos, P.: Joint unsupervised coreference resolution with markov logic. In: Proceedings of EMNLP, pp. 650–659 (2008)Google Scholar
  24. 24.
    Song, Y., Jiang, J., Zhao, W.X., Li, S., Wang, H.: Joint learning for coreference resolution with markov logic. In: Proceedings of the 2012 Joint Conference on Empirical Methods in Natural Language Processing and Computational Natural Language Learning, pp. 1245–1254. ACL (2012)Google Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 2013

Authors and Affiliations

  • Kazeto Yamamoto
    • 1
  • Naoya Inoue
    • 1
  • Yotaro Watanabe
    • 1
  • Naoaki Okazaki
    • 1
  • Kentaro Inui
    • 1
  1. 1.Tohoku UniversityJapan

Personalised recommendations