Advertisement

Spectral Learning of Sequence Taggers over Continuous Sequences

  • Adrià Recasens
  • Ariadna Quattoni
Part of the Lecture Notes in Computer Science book series (LNCS, volume 8188)

Abstract

In this paper we present a spectral algorithm for learning weighted finite-state sequence taggers (WFSTs) over paired input-output sequences, where the input is continuous and the output discrete. WFSTs are an important tool for modelling paired input-output sequences and have numerous applications in real-world problems. Our approach is based on generalizing the class of weighted finite-state sequence taggers over discrete input-output sequences to a class where transitions are linear combinations of elementary transitions and the weights of the linear combination are determined by dynamic features of the continuous input sequence. The resulting learning algorithm is efficient and accurate.

Keywords

Feature Function Hide Markov Model Sequence Tagger Gesture Recognition Continuous Sequence 
These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.

References

  1. 1.
    Asuncion, A., Newman, D.J.: UCI machine learning repository (2007), http://www.ics.uci.edu/~mlearn/MLRepository.html
  2. 2.
    Anandkumar, A., Hsu, D., Kakade, S.M.: A method of moments for mixture models and hidden markov models. CoRR abs/1203 0683 (2012)Google Scholar
  3. 3.
    Bailly, R.: Quadratic weighted automata: Spectral algorithm and likelihood maximization. Journal of Machine Learning Research (2011)Google Scholar
  4. 4.
    Bailly, R., Denis, F., Ralaivola, L.: Grammatical inference as a principal component analysis problem. In: Proc. ICML (2009)Google Scholar
  5. 5.
    Balle, B., Quattoni, A., Carreras, X.: A spectral learning algorithm for finite state transducers. In: Gunopulos, D., Hofmann, T., Malerba, D., Vazirgiannis, M. (eds.) ECML PKDD 2011, Part I. LNCS, vol. 6911, pp. 156–171. Springer, Heidelberg (2011)CrossRefGoogle Scholar
  6. 6.
    Balle, B., Quattoni, A., Carreras, X.: Local loss optimization in operator models: A new insight into spectral learning. In: Proceedings of ICML, pp. 1879–1886 (2012)Google Scholar
  7. 7.
    Blei, D.M., Ng, A.Y., Jordan, M.I., Lafferty, J.: Latent dirichlet allocation. Journal of Machine Learning Research 3, 2003 (2003)Google Scholar
  8. 8.
    Boots, B., Siddiqi, S., Gordon, G.: Closing the learning planning loop with predictive state representations. I. J. Robotic Research (2011)Google Scholar
  9. 9.
    Boots, B., Gordon, G.J.: An online spectral learning algorithm for partially observable nonlinear dynamical systems. In: Proceedings of the 25th National Conference on Artificial Intelligence (2001)Google Scholar
  10. 10.
    Bosch, A., Zisserman, A., Muñoz, X.: Scene classification via pLSA. In: Leonardis, A., Bischof, H., Pinz, A. (eds.) ECCV 2006, Part IV. LNCS, vol. 3954, pp. 517–530. Springer, Heidelberg (2006)CrossRefGoogle Scholar
  11. 11.
    Chang, J.T.: Full reconstruction of markov models on evolutionary trees: Identifiability and consistency. Mathematical Biosciences 137, 51–73 (1996)MathSciNetzbMATHCrossRefGoogle Scholar
  12. 12.
    Clark, A.: Partially supervised learning of morphology with stochastic transducers. In: Proc. of NLPRS, pp. 341–348 (2001)Google Scholar
  13. 13.
    Eisner, J.: Parameter estimation for probabilistic finite-state transducers. In: Proc. of ACL, pp. 1–8 (2002)Google Scholar
  14. 14.
    Gao, J., Johnson, M.: A comparison of Bayesian estimators for unsupervised Hidden Markov Model POS taggers. In: Proceedings of EMNLP, pp. 344–352 (2008)Google Scholar
  15. 15.
    Hsu, D., Kakade, S.M., Zhang, T.: A spectral algorithm for learning hidden markov models. In: Proc. of COLT (2009)Google Scholar
  16. 16.
    Jaeger, H.: Observable operator models for discrete stochastic time series. Neural Computation 12, 1371–1398 (2000)CrossRefGoogle Scholar
  17. 17.
    Jaeger, H.: Characterizing distributions of stochastic processes by linear operators. Tech. Rep. 62, German National Research Center for Information Technology (1999)Google Scholar
  18. 18.
    Jaeger, H.: Modeling and learning continuous-valued stochastic processes with ooms. Tech. Rep. 102, German National Research Center for Information Technology (2001)Google Scholar
  19. 19.
    Luque, F., Quattoni, A., Balle, B., Carreras, X.: Spectral learning in non-deterministic dependency parsing. In: EACL (2012)Google Scholar
  20. 20.
    Morency, L.P., Quattoni, A., Darrell, T.: Latent-dynamic discriminative models for continuous gesture recognition. In: CVPR (2007)Google Scholar
  21. 21.
    Mossel, E., Roch, S.: Learning nonsingular phylogenies and hidden markov models. In: Proc. of STOC (2005)Google Scholar
  22. 22.
    Quattoni, A., Wang, S., Morency, L., Collins, M., Darrell, T.: Hidden-state conditional random fields. Pattern Analysis and Machine Intelligence (2007)Google Scholar
  23. 23.
    Shibata, C., Yoshinaka, R.: Marginalizing out transition probabilities for several subclasses of pfas. In: JMLR Workshop and Conference Proceedings, ICGI 2012, vol. 21, pp. 259–263 (2012)Google Scholar
  24. 24.
    Siddiqi, S.M., Boots, B., Gordon, G.J.: Reduced-Rank Hidden Markov Models. In: Proc. AISTATS, pp. 741–748 (2010)Google Scholar
  25. 25.
    Siddiqi, S., Boots, B., Gordon, G.J.: Reduced-rank hidden Markov models. In: Proceedings of the Thirteenth International Conference on Artificial Intelligence and Statistics, AISTATS 2010 (2010)Google Scholar
  26. 26.
    Song, L., Boots, B., Siddiqi, S.M., Gordon, G.J., Smola, A.J.: Hilbert space embeddings of hidden Markov models. In: Proc. 27th Intl. Conf. on Machine Learning, ICML (2010)Google Scholar
  27. 27.
    Song, L., Huang, J., Smola, A., Fukumizu, K.: Hilbert space embeddings of conditional distributions with applications to dynamical systems (2009)Google Scholar
  28. 28.
    Wang, S.B., Quattoni, A., Morency, L.P., Demirdjian, D.: Hidden conditional random fields for gesture recognition. In: Proceedings of the 2006 IEEE Computer Society Conference on Computer Vision and Pattern Recognition, CVPR 2006, vol. 2, pp. 1521–1527 (2006)Google Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 2013

Authors and Affiliations

  • Adrià Recasens
    • 1
  • Ariadna Quattoni
    • 1
  1. 1.Universitat Politècnica de CatalunyaBarcelonaSpain

Personalised recommendations