Advertisement

Network of Recurrent Neural Networks: Design for Emergence

  • Chaoming Wang
  • Yi Zeng
Conference paper
Part of the Lecture Notes in Computer Science book series (LNCS, volume 11302)

Abstract

Emergence plays an important role in Recurrent Neural Networks (RNNs). In order to design for emergence, we qualitatively and quantitatively design the recurrent neural network structures from the perspective of systems theory. From the qualitative viewpoint, we introduce two methodologies (aggregation and specialization) from systems theory to design the novel neural structure, and we name it as “Network Of Recurrent neural networks” (NOR). In NOR, RNNs are viewed as the high-level neurons and are used to build the high-level layers. Experiments on three predictive tasks show that under the same number of parameters, the implemented NOR models get superior performances than conventional RNN structures (e.g., vanilla RNN, LSTM and GRU). More importantly, from the quantitative perspective, we introduce an information-theoretical framework to quantify the information dynamics in recurrent neural structures. And the evaluation results show that several NOR models achieve similar or better emergent information processing capabilities compared with LSTM.

Keywords

Recurrent Neural Networks Systems theory Emergence 

References

  1. 1.
    Arthur, W.B.: On the evolution of complexity. In: Cowan, G.A., Pines, D., Meltzer, D.E. (eds.) Complexity: Metaphors, Models, and Reality. Advanced Book Classics, pp. 65–81. Westview Press, Cambridge (1999). Chapter 5Google Scholar
  2. 2.
    Bahdanau, D., Cho, K., Bengio, Y.: Neural machine translation by jointly learning to align and translate. arXiv preprint arXiv:1409.0473 (2014)
  3. 3.
    Bertschinger, N., Natschläger, T.: Real-time computation at the edge of chaos in recurrent neural networks. Neural Comput. 16(7), 1413–1436 (2004)CrossRefGoogle Scholar
  4. 4.
    Boedecker, J., Obst, O., Lizier, J.T., Mayer, N.M., Asada, M.: Information processing in echo state networks at the edge of chaos. Theor. Biosci. 131(3), 205–213 (2012)CrossRefGoogle Scholar
  5. 5.
    Cho, K., et al.: Learning phrase representations using RNN encoder-decoder for statistical machine translation. arXiv preprint arXiv:1406.1078 (2014)
  6. 6.
    Dessalles, J.L., Müller, J.P., Phan, D.: Emergence in multi-agent systems: conceptual and methodological issues. In: Phan, D., Amblard, F. (eds.) Agent-based Modelling and Simulation in the Social and Human Sciences, pp. 327–355. The Bardwell Press, Oxford (2007)Google Scholar
  7. 7.
    Elman, J.L.: Finding structure in time. Cogn. Sci. 14(2), 179–211 (1990)CrossRefGoogle Scholar
  8. 8.
    Fromm, J.: The Emergence of Complexity. Kassel University Press, Kassel (2004)Google Scholar
  9. 9.
    Hochreiter, S., Schmidhuber, J.: Long short-term memory. Neural Comput. 9(8), 1735–1780 (1997)CrossRefGoogle Scholar
  10. 10.
    Holland, J.H.: Emergence: From Chaos to Order. OUP, Oxford (2000)zbMATHGoogle Scholar
  11. 11.
    Huang, Z., Xu, W., Yu, K.: Bidirectional LSTM-CRF models for sequence tagging. arXiv preprint arXiv:1508.01991 (2015)
  12. 12.
    Kingma, D.P., Ba, J.: Adam: a method for stochastic optimization. arXiv preprint arXiv:1412.6980 (2014)
  13. 13.
    Le, Q.V., Jaitly, N., Hinton, G.E.: A simple way to initialize recurrent networks of rectified linear units. arXiv preprint arXiv:1504.00941 (2015)
  14. 14.
    LeCun, Y., Bottou, L., Bengio, Y., Haffner, P.: Gradient-based learning applied to document recognition. Proc. IEEE 86(11), 2278–2324 (1998)CrossRefGoogle Scholar
  15. 15.
    Lehn, J.M.: Towards complex matter: supramolecular chemistry and self-organization. Eur. Rev. 17(2), 263–280 (2009)CrossRefGoogle Scholar
  16. 16.
    Li, X., Roth, D.: Learning question classifiers. In: Proceedings of the 19th International Conference on Computational Linguistics-Volume 1, pp. 1–7. Association for Computational Linguistics (2002)Google Scholar
  17. 17.
    Lizier, J.T., Prokopenko, M., Zomaya, A.Y.: A framework for the local information dynamics of distributed computation in complex systems. In: Prokopenko, M. (ed.) Guided Self-Organization: Inception. ECC, vol. 9, pp. 115–158. Springer, Heidelberg (2014).  https://doi.org/10.1007/978-3-642-53734-9_5CrossRefGoogle Scholar
  18. 18.
    Mele, C., Pels, J., Polese, F.: A brief review of systems theories and their managerial applications. Serv. Sci. 2, 126–135 (2010)CrossRefGoogle Scholar
  19. 19.
    Mitchell, M.: Complexity: A guided Tour. Oxford University Press, New York (2009)zbMATHGoogle Scholar
  20. 20.
    Nicolis, G., Prigogine, I.: Self-organization in Nonequilibrium Systems: From Dissipative Structures to Order Through Fluctuations. Wiley (1977)Google Scholar
  21. 21.
    Pascanu, R., Mikolov, T., Bengio, Y.: On the difficulty of training recurrent neural networks. In: International Conference on Machine Learning, pp. 1310–1318 (2013)Google Scholar
  22. 22.
    Pennington, J., Socher, R., Manning, C.: Glove: global vectors for word representation. In: Proceedings of the 2014 Conference on Empirical Methods in Natural Language Processing (EMNLP), pp. 1532–1543 (2014)Google Scholar
  23. 23.
    Kárný, M., Warwick, K., Kůrková, V.: Recurrent neural networks: some systems-theoretic aspects. In: Kárnỳ, M., Warwick, K., Kůrková, V. (eds.) Dealing with Complexity, pp. 1–12. Springer, London (1998).  https://doi.org/10.1007/978-1-4471-1523-6_1CrossRefzbMATHGoogle Scholar
  24. 24.
    Tjong Kim Sang, E.F., De Meulder, F.: Introduction to the CoNLL-2003 shared task: language-independent named entity recognition. In: Proceedings of the Seventh Conference on Natural Language Learning at HLT-NAACL, vol. 4, pp. 142–147. Association for Computational Linguistics (2003)Google Scholar
  25. 25.
    Von Bertalanffy, L.: General System Theory. G. Braziller, New York (1968)Google Scholar
  26. 26.
    Zaremba, W., Sutskever, I., Vinyals, O.: Recurrent neural network regularization. arXiv preprint arXiv:1409.2329 (2014)

Copyright information

© Springer Nature Switzerland AG 2018

Authors and Affiliations

  1. 1.Institute of Automation, Chinese Academy of SciencesBeijingChina
  2. 2.University of Chinese Academy of SciencesBeijingChina
  3. 3.Center for Excellence in Brain Science and Intelligence TechnologyChinese Academy of SciencesShanghaiChina

Personalised recommendations