Advertisement

Improving the Classification Performance of Liquid State Machines Based on the Separation Property

  • Emmanouil Hourdakis
  • Panos Trahanias
Part of the IFIP Advances in Information and Communication Technology book series (IFIPAICT, volume 363)

Abstract

Liquid State Machines constitute a powerful computational tool for carrying out complex real time computations on continuous input streams. Their performance is based on two properties, approximation and separation. While the former depends on the selection of class functions for the readout maps, the latter needs to be evaluated for a particular liquid architecture. In the current paper we show how the Fisher’s Discriminant Ratio can be used to effectively measure the separation of a Liquid State Machine. This measure is then used as a fitness function in an evolutionary framework that searches for suitable liquid properties and architectures in order to optimize the performance of the trained readouts. Evaluation results demonstrate the effectiveness of the proposed approach.

Keywords

Liquid State Machines Separation Fisher Discriminant Ratio 

References

  1. 1.
    Auer, P., Burgsteiner, H., Maass, W.: A learning rule for very simple universal approximators consisting of a single layer. Neural Nets 21, 786–795 (2008)CrossRefGoogle Scholar
  2. 2.
    Burgsteiner, H., Kroll, M., Leopold, A., Steinbauer, G.: Movement prediction from real-world images using a liquid state machine. Applied Intelligence 27(2), 99–109 (2007)CrossRefGoogle Scholar
  3. 3.
    Dockendorf, K.P., Park, I., He, P., Principe, J.C., DeMarse, T.B.: Liquid state machines and cultured cortical networks: The separation property. BioSystems 95(2), 90–97 (2009)CrossRefGoogle Scholar
  4. 4.
    Duda, R.O., Hart, P.E., Stork, D.G.: Pattern Classification. Wiley, Chichester (2000)Google Scholar
  5. 5.
    Fernando, C., Sojakka, S.: Pattern recognition in a bucket. Advances in ALife, 588–597 (2003)Google Scholar
  6. 6.
    Fogel, D.B.: An introduction to simulated evolutionary optimization. IEEE Trans. Neural Networks 5, 3–14 (1994)CrossRefGoogle Scholar
  7. 7.
    Goodman, E., Ventura, D.: Spatiotemporal pattern recognition via liquid state machines. In: Intl. Joint Conf. Neural Networks, IJCNN, pp. 3848–3853 (2006)Google Scholar
  8. 8.
    Legenstein, R., Maass, W.: Edge of chaos and prediction of computational performance for neural circuit models. Neural Nets 20(3), 323–334 (2007)MATHCrossRefGoogle Scholar
  9. 9.
    Maass, W., Natschlaeger, T., Markram, H.: Real-time computing without stable states: A new framework for neural computation based on perturbations. Neural Computation 14(11), 2531–2560 (2002)MATHCrossRefGoogle Scholar
  10. 10.
    Natschlager, T., Markram, H., Maass, W.: Computer models and analysis tools for neural microcircuits. A practical guide, pp. 123–138 (2003) Google Scholar
  11. 11.
    Rieke, F.: Spikes: Exploring the neural code. MIT Press, Cambridge (1999)Google Scholar
  12. 12.
    Rumelhart, D., Hinton, G., Williams, R.: Learning Internal Representations by Error Propagation. In: Parallel Distributed Processing: Explorations in the Microstructure of Cognition, vol. 1. MIT Press, Cambridge (1986)Google Scholar
  13. 13.
    Stein: Some models of neuronal variability. Biophysical Journal 7, 37–68 (1967)CrossRefGoogle Scholar
  14. 14.
    Verstraeten, D., Schrauwen, B., Stroobandt, D.: Isolated word recognition with the liquid state machine: a case study. In: 13th European Symp. Artificial Neural Nets (ESANN), vol. 95(6), pp. 435–440 (2005)Google Scholar
  15. 15.
    Yao, X.: Evolving artificial neural networks. Proc. of the IEEE 87(9), 1423–1447 (1999)CrossRefGoogle Scholar
  16. 16.
    Norton, D., Ventura, D.: Improving the performance of liquid state machines through iterative refinement of the reservoir. Neurocomputing 73, 2893–2904 (2010)CrossRefGoogle Scholar
  17. 17.
    Kok, S.: Liquid State Machine Optimization, Master Thesis, Utrecht University (2007)Google Scholar
  18. 18.
    Matser, J.: Structured liquids in liquid state machines, Master Thesis, Utrecht University (2010)Google Scholar
  19. 19.
    Fisher, R.A.: “The Use of Multiple Measurements in Taxonomic Problems” (PDF). Annals of Eugenics 7(2), 179–188 (1936)CrossRefGoogle Scholar

Copyright information

© International Federation for Information Processing 2011

Authors and Affiliations

  • Emmanouil Hourdakis
    • 1
  • Panos Trahanias
    • 1
  1. 1.Institute of Computer ScienceFoundation for Research and Technology – Hellas (FORTH)HeraklionGreece

Personalised recommendations