Advertisement

Neural Spatial Interaction Models: Network Training, Model Complexity and Generalization Performance

  • Manfred M. Fischer
Conference paper
Part of the Lecture Notes in Computer Science book series (LNCS, volume 7974)

Abstract

Spatial interaction models approximate mean interaction frequencies between origin and destination locations by using origin-specific, destination-specific and spatial separation information. The focus is on models that are based on the theory of feedforward neural networks. This contribution considers the functional form of neural spatial interaction models, including the specification of the activation functions, and discusses the problem of network training within a maximum likelihood framework that involves the solution of a non-linear optimization problem. This requires the evaluation of the log-likelihood function with respect to the network parameters. Overfitting is a problem that is likely to occur in neural spatial interaction models. To avoid this problem the contribution recommends controlling the model complexity either by regularization or by early stopping in network training. A bootstrapping pairs approach with replacement may be adopted to evaluate the generalization performance of the models.

Keywords

Neural spatial interaction models network training gradient based search model complexity regularization bootstrapping generalization performance 

Preview

Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.

References

  1. Baldi, P., Chauvin, Y.: Temporal evolution of generalization during learning in linear networks. Neural Computation 3(4), 589–603 (1991)CrossRefGoogle Scholar
  2. Bergkvist, E.: Forecasting interregional freight flows by gravity models. Jahrbuch für Regionalwissenschaft 20(2), 133–148 (2000)Google Scholar
  3. Bergkvist, E., Westin, L.: Estimation of gravity models by OLS estimation, NLS estimation, Poisson and neural network specifications. CERUM Regional Dimensions, WP-6 (1997)Google Scholar
  4. Bishop, C.M.: Pattern recognition and machine learning. Springer, New York (2006)zbMATHGoogle Scholar
  5. Bishop, C.M.: Neural networks for pattern recognition. Clarendon Press, Oxford (1995)Google Scholar
  6. Black, W.P.: Spatial interaction modeling using artificial neural networks. Journal of Transportation Geography 3(3), 159–166 (1995)CrossRefGoogle Scholar
  7. Cybenko, G.: Approximation by superpositions of a sigmoidal function. Mathematics of Control Signals and Systems 2(4), 303–314 (1989)MathSciNetzbMATHCrossRefGoogle Scholar
  8. Efron, B.: The jackknife, the bootstrap and other resampling plans. Society for Industrial and Applied Mathematics, Philadelphia (1982)CrossRefGoogle Scholar
  9. Finnoff, W.: Complexity measures for classes of neural networks with variable weight bounds. In: Proceedings of the International Geoscience and Remote Sensing Symposium, IGARSS 1994, vol. 4, pp. 1880–1882. IEEE Press, Piscataway (1991)Google Scholar
  10. Fischer, M.M.: Spatial interaction models. In: Warf, B. (ed.) Encyclopedia of Geography, pp. 2645–2647. Sage Publications, London (2010)Google Scholar
  11. Fischer, M.M.: Learning in neural spatial interaction models: A statistical perspective. Journal of Geographical Systems 4(3), 287–299 (2002)CrossRefGoogle Scholar
  12. Fischer, M.M.: Methodological challenges in neural spatial interaction modelling: The issue of model selection. In: Reggiani, A. (ed.) Spatial Economic Science: New Frontiers in Theory and Methodology, pp. 89–101. Springer, Heidelberg (2000)CrossRefGoogle Scholar
  13. Fischer, M.M., Gopal, S.: Artificial neural networks. A new approach to modelling interregional telecommunication flows. Journal of Regional Science 34(4), 503–527 (1994)CrossRefGoogle Scholar
  14. Fischer, M.M., Leung, Y.: A genetic-algorithm based evolutionary computational neural network for modelling spatial interaction data. The Annals of Regional Science 32(3), 437–458 (1998)CrossRefGoogle Scholar
  15. Fischer, M.M., Reismann, M.: Evaluating neural spatial interaction modelling by bootstrapping. Networks and Spatial Economics 2(3), 255–268 (2002a)Google Scholar
  16. Fischer, M.M., Reismann, M.: A methodology for neural spatial interaction modeling. Geographical Analysis 34(3), 207–228 (2002b)Google Scholar
  17. Fischer, M.M., Hlavácková-Schindler, K., Reismann, M.: A global search procedure for parameter estimation in neural spatial interaction modelling. Papers in Regional Science 78(2), 119–134 (1999)CrossRefGoogle Scholar
  18. Fischer, M.M., Reismann, M., Hlavácková-Schindler, K.: Neural network modelling of constrained spatial interaction flows: Design, estimation and performance issues. Journal of Regional Science 43(1), 35–61 (2003)CrossRefGoogle Scholar
  19. Funahashi, K.: On the approximate realization of continuous mappings by neural networks. Neural Networks 2(3), 183–192 (1989)CrossRefGoogle Scholar
  20. Haykin, S.: Neural networks. A comprehensive foundation. Macmillan College Publishing Company, New York (1994)zbMATHGoogle Scholar
  21. Hornik, K., Stinchcombe, M., White, H.: Multilayer feedforward networks are universal approximators. Neural Networks 2(5), 359–368 (1989)CrossRefGoogle Scholar
  22. Kullback, S., Leibler, R.A.: On information and sufficiency. Annals of Mathematical Statistics 22(1), 78–86 (1951)MathSciNetCrossRefGoogle Scholar
  23. Mozolin, M., Thill, J.-C., Usery, E.L.: Trip distribution forecasting with multilayer perceptron neural networks: A critical evaluation. Transportation Research B 34(1), 53–73 (2000)CrossRefGoogle Scholar
  24. Nijkamp, P., Reggiani, A., Tritapepe, A.: Modelling intra-urban transport flows in Italy. TRACE Discussion Papers TI 96-60/5. Tinbergen Institute, The Netherlands (1996)Google Scholar
  25. Nocedal, J., Wright, S.J.: Numerical optimization. Springer, New York (1999)zbMATHCrossRefGoogle Scholar
  26. Openshaw, S.: Modelling spatial interaction using a neural net. In: Fischer, M.M., Nijkamp, P. (eds.) Geographic Information Systems, Spatial Modelling, and Policy Evaluation, pp. 147–164. Springer, Heidelberg (1993)CrossRefGoogle Scholar
  27. Press, W.H., Teukolky, S.A., Vetterling, W.T., Flannery, B.P.: Numerical recipes in C. The art of scientific computing, 2nd edn. Cambridge University Press, Cambridge (1992)zbMATHGoogle Scholar
  28. Reggiani, A., Tritapepe, T.: Neural networks and logit models applied to commuters’ mobility in the metropolitan area of Milan. In: Himanen, V., Nijkamp, P., Reggiani, A. (eds.) Neural Networks in Transport Applications, pp. 111–129. Ashgate, Aldershot (2000)Google Scholar
  29. Rumelhart, D.E., Durbin, R., Golden, R., Chauvin, Y.: Backpropagation: The basic theory. In: Chauvin, Y., Rumelhart, D.E. (eds.) Backpropagation: Theory, Architectures and Applications, pp. 1–34. Lawrence Erlbaum Associates, Hillsdale (1995)Google Scholar
  30. Rumelhart, D.E., Hinton, G.E., Williams, R.J.: Learning internal representations by error propagation. In: Rumelhart, D.E., McClelland, J.L. (eds.) Parallel Distributed Processing: Explorations in the Microstructure of Cognition, pp. 318–362. MIT Press, Cambridge (1986)Google Scholar
  31. Thill, J.-C., Mozolin, M.: Feedforward neural networks for spatial interaction: Are they trustworthy forecasting tools? In: Reggiani, A. (ed.) Spatial Economic Science: New Frontiers in Theory and Methodology, pp. 355–381. Springer, Heidelberg (2000)CrossRefGoogle Scholar
  32. Tibshirani, R.: Regression shrinkage and selection via the lasso. Journal of the Royal Statistical Society B 58(1), 267–288 (1996)MathSciNetzbMATHGoogle Scholar
  33. Weigend, A.S., Rumelhart, D.E., Huberman, B.A.: Generalization by weight elimination with application to forecasting. In: Lippman, R., Moody, J., Touretzky, D. (eds.) Advances in Neural Information Processing Systems, vol. 3, pp. 875–882. Morgan Kaufmann, San Mateo (1991)Google Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 2013

Authors and Affiliations

  • Manfred M. Fischer
    • 1
  1. 1.Vienna University of Economics and BusinessAustria

Personalised recommendations