Die Architektur- und Werteinstellungsproblematik der Parameter Neuronaler Netze

  • Walter Frisch
Conference paper

Zusammenfassung

Seit Ende der achtziger Jahre werden Neuronale Netze verstärkt zur Lösung ökonomischer Probleme eingesetzt. Der vorhegende Überblick diskutiert den Charakter der Parameter in der Architektur und der Werteinstellung Neuronaler Netze und gibt einen Überblick über bereits bestehende Verfahren zur günstigen Voreinstellung und Konfigurierung.

This is a preview of subscription content, log in to check access.

Preview

Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.

Literatur

  1. [1]
    Baetge, J., Möglichkeiten der Früherkennung negativer Unternehmensentwicklungen mit Hilfe statistischer Jahresabschlußanalysen, in: Zeitschrift für betriebswirtschaftliche Forschung 41, S. 792 – 810.Google Scholar
  2. [2]
    Barna, G. and Kaski, K., Choosing Optimal Network Structure, Proc. of the Internat. Neural Network Conf., Paris (1990)Google Scholar
  3. [3]
    Barto, A. G. and Singh, S. P., On the Computational Economics of Reinforcement Learning, Connectionist Models, Proceedings of the 1990 Summer School (1990)Google Scholar
  4. [4]
    Baum, E. B., Haussler, D., What Size Net Gives Valid Generalization? in: Advance in Neural Information Processing Systems 1, David Touretzky (Ed.), 1989Google Scholar
  5. [5]
    Bilbro, G., Mann, R., Miller, T.K., Snyder, W.E., Van den Bout, D.E., White, M., Optimization by Mean Field Annealing, in: Advance in Neural Information Processing Systems 1, David Touretzky (Ed.), 1989Google Scholar
  6. [6]
    Braun, H., Weisbrod, J., Evolving Neural Networks for Application Oriented Problems, Proceedings of the Conference on Evolutionary Programming, San Diego (1993)Google Scholar
  7. [7]
    Chauvin, Y., A Back-Propagation Algorithm with Optimal Use of Hidden Units, in: Advance in Neural Information Processing Systems 1, David Touretzky (Ed.), 1989Google Scholar
  8. [8]
    Cohen, M.A., Grossberg, S., Absolute Stability of Global Pattern Formation and Parallel Memory Storage by Competitive Neural Networks, IEEE Trans, on Systems, Man and Cybernetics, SMC-13, No. 5, Sept/Oct 1983, 815–826 (1983)Google Scholar
  9. [9]
    Crowder, R. Scott, Predicting the Mackey-Glass Timeseries With Cascade-Correlation Learning, Connectionist Models, Proceedings of the 1990 Summer School (1990)Google Scholar
  10. [10]
    Engle, R.F., Yoo, B.S., Forecasting and Testing in Co-Integrated Systems, Journal of Econometrics 35, 143–159 (1987)CrossRefGoogle Scholar
  11. [11]
    Geyer, A., Geyer-Schulz, A., Taudes, A., An Expert System for Time-Series Analysis, Statistik, Informatik und Ökonomie, Wolfgang Janko (Hrsg.), Springer-Verlag Berlin Heidelberg 1988Google Scholar
  12. [12]
    Geyer-Schulz, A., Fuzzy Rule-Based Expert Systems and Genetic Machine Learning, to appear (1993)Google Scholar
  13. [13]
    Geyer-Schulz, A., On the Specification of Fuzzy Data in Management, in Bandemer, S. 105–110 (1993)Google Scholar
  14. [14]
    Goldberg, D.E., Optimal Initial Population Size for Binary-Coded Genetic Algorithms, Technical Report, TCGA No. 85001, The Clearinghouse for Genetic Algorithms, University of Alabama (1985)Google Scholar
  15. [15]
    Goldberg, D.E., Genetic Algorithms in Search, Optimization and Machine Learning, Reading, PA: Lawrence Erlbaum (1985)Google Scholar
  16. [16]
    Grefenstette, J.J. (Ed.), Genetic Algorithms and Their Applications: Proceedings of the Second Int. Conf. on Genetic Algorithms, Cambridge, MA: Lawrence Erlbaum (1987)Google Scholar
  17. [17]
    Hansson, O., Mayer, A., Probabilistic Heuristic Estimates, Anals of Mathematics and Artificial Intelligence, 2, D.J. Hand (ed.) (1990)Google Scholar
  18. [18]
    Harp, S., Sainad, T. and Guha, A., Towards the Genetic Synthesis of Neural Networks, in: Proc. Third Intern. Conf. Genetic Algorithms, Morgan Kaufmann, San Mateo, CA (1989)Google Scholar
  19. [19]
    Hecht-Nielsen, R., Neurocomputing, Reading: Addison-Wesley, 1990Google Scholar
  20. [20]
    Hecht-Nielsen, R., On the Algebraic Structure of Feedforward Network Weight Spaces, in: Advanced Neural Computers, R. Eckmiller (Ed.) (1990)Google Scholar
  21. [21]
    Hergert, F., Finnoff, W., Zimmermann, H.G. A Comparison of Weight Elimination Methods for Reducing Complexity in Neural Networks, Siemens AG, Corporate Research and Development, Otto-Hahn-Ring 6., 8000 Munchen 83 (1992)Google Scholar
  22. [22]
    Hertz, J., Krogh, A., Palmer, R.G. Introduction to the Theory of Neural Networks, Addison-Wesley, Redwood City, California (1991)Google Scholar
  23. [23]
    Huberman, B., Rumelhart, D., Weigend, A. (Huberman et al., 1991): Generalisation by Weight Elimination with Applications to Forecasting, in: Lippmann, R.P., Moody J.(Eds.), Advances in Neural Information Processing III, Morgan Kaufmann (1991)Google Scholar
  24. [24]
    Huberman, B. A., Rumelhart, D. E. and Weigend A. S., Back-Propagation, Weight-Elimination and Time Series Prediction, Connectionist Models, Proceedings of the 1990 Summer School (1990)Google Scholar
  25. [25]
    Janko, W.H., Probalistic Algorithms and the Efficient Use of Statistical Models of Decision in their Design, in: Methods of Operations Research, Henn et al. (eds.), Vol. 36, Athenäum/Hain/Scriptor/Hanstein, Königstein/Ts. (1980)Google Scholar
  26. [26]
    Janko, W.H., Taudes, A., Frisch, W., Simultane Alternativensuche und Datenpräzisierung, Diskussionspapiere zum Tätigkeitsfeld Informationsverarbeitung und Informationswirtschaft Nr. 13, H.R. Hansen, W.H. Janko (Hrsg.) (1993)Google Scholar
  27. [27]
    Janko, W.H., Searching in the Best Offer When the Distribution of Offers Is Truly Unknown, in: Contributions to Econometrics and Statistics Today, Schneeweiß und Strecker (eds.), Springer, Berlin/Heidelberg (1985)Google Scholar
  28. [28]
    Kenney, N., Lippmann, R.P., A Comparative Study of the Practical Cha-ractersitics of Neural Network and Conventional Pattern Classifiers, in: Advances in Neural Information Processing Systems 3, R.P. Lippmann, J.E. Moody, D.S. Touretzky (eds.) (1989)Google Scholar
  29. [29]
    Kolen, J. F., Pollack, J.B., Back Propagation is Sensitive to Initial Conditions, in: Advances in Neural Information Processing Systems 3, R.P. Lippmann, J.E. Moody and D.S. Touretzky (eds.), 1986Google Scholar
  30. [30]
    Koza, J.R., The Genetic Programming Paradigm: Genetically Breeding Populations of Computer Programs to Solve Problems, in: Dynamic, Genetic, and Chaotic Programming, B. Soucek (Ed.), Wiley & Sons (1992)Google Scholar
  31. [31]
    Kruse, R., Meyer, K.D., Statistics with Vague Data, D. Reidel Publishing Company, Dordrecht (1987)CrossRefGoogle Scholar
  32. [32]
    Kushner, H., Asymptotic Global Behavior for Stochastic Approximations and Diffusions With Slowly Decreasing Noise Effects: Global Minimization via Monte Carlo, SIAM, Journal of Applied Mathematics, 47, 169–185 (1987)Google Scholar
  33. [33]
    Le Cun, Y., Denker, J.S., Soila, S.A. Optimal Brain Damage, in: Touretzky, D.S. (Ed.): Advances in Neural Information Processing Systems II, Morgan Kaufmann, San Mateo, S. 598 ff. (1990)Google Scholar
  34. [34]
    MacQueen, J.B., Optimal Policies for a Class of Search and Avaluation Problems, Management Science, Vol. 10, No. 4 (1964)CrossRefGoogle Scholar
  35. [35]
    Matthes, R., Zinsprognosen: Fehlerkorrekturmodelle vs. Neuronaler Netze, Referat 4. Karlsruher Ökonometrie Workshop, (1993)Google Scholar
  36. [36]
    Miller M., Das Optimieren von Neuronalen Netzen für den Einsatz zur Prognose in der Ökonomie, Referat 4. Karlsruher Ökonomie-Workshop, (1993)Google Scholar
  37. [37]
    Miller, G., Todd, P. and Hedge, S., Designing Neural Networks Using Genetic Algorithms, in: Proc. Third Internat. Conf. Genetic Algorithms, Morgan Kaufmann, San Mateo, CA (1989)Google Scholar
  38. [38]
    Moody, J.E., The Effective Number of Parameters: An Analysis of Generalization and Regularization in Nonlinear Learning Systems. Appears in J.E. Moody, S.J. Hanson and R.P. Lippmann (editors): Advances in Neural Information Processing Systems 4, Morgan Kaufmann Publishers, San Mateo, C.A (1992)Google Scholar
  39. [39]
    Mühlenbein, H., Kindermann, J., The Dynamics of Evolution and Learning — Towards Genetic Neural Networks, Connectionism in Perspective, Pfeiffer ed. (1989)Google Scholar
  40. [40]
    Natter, M., Lukanowicz, M., Ein konnexionistischer Ansatz zur Prognose und Steuerung von Ablaufschadstoffen einer Kläranlage, in: K.W. Hausmann, M. Bachens, M. Janke, W.E. Katzenberger, A. Maruser (eds.), Operations Research Proceedings, Springer Verlag Berling, S. 95–102 (1992)Google Scholar
  41. [41]
    Odom, M.D., Sharda, R. A. Neural Network Model for Bankruptcy Prediction, IJCNN International Joint Conference on Neural Networks, San Diego, California, June 17–21, 1990, Vol. II, S. II-63 ff. (1990)Google Scholar
  42. [42]
    Radermacher, F.J., Das Paradigma. Neuronales Netz/Konnektionismus: Einige Anmerkungen und Hinweise zu Anwendungen, Referat 4. Karlsruher Ökonomie-Workshop (1993)Google Scholar
  43. [43]
    Rehkugler, H., Poddig, Th. Statistische Methoden versus Künstliche Neuronale Netzwerke zur Aktienkursprognose, — Eine vergleichende Studie -, Bamberger Betriebswirtschaftliche Beiträge, Nr. 73/1990, Otto-Friedrich-Universität Bamberg (1990)Google Scholar
  44. [44]
    Rehkugler, H., Poddig, Th. Entwicklung leistungsfähiger Prognosesysteme auf Basis Künstlicher Neuronaler Netzwerke am Beispiel des Dollars, — Eine Fallstudie -, Bamberger Betriebswirtschaftliche Beiträge, Nr. 76/1990, Otto-Friedrich-Universität Bamberg (1990)Google Scholar
  45. [45]
    Rehkugler, H., Poddig, Th. Neuronale Netze im Bankbetrieb, in: Die Bank, 7, S 413 ff. (1992)Google Scholar
  46. [46]
    Reski, Th.: Paralelle Simulation Neuronaler Netzwerke, in: Transputer Anwender Treffen, S. 212 ff. (1991)Google Scholar
  47. [47]
    Rohwer, R., Time Trials on Second-Order and Variable-Learning-Rate Algorithms, in: Advances in Neural Information Processing Systems 3, R.P. Lippmann, J.E. Moody, D.S. Touretzky (eds.) (1989)Google Scholar
  48. [48]
    Rossen, M.L., Closed-Form Inversion of Backpropagation Networks: Theory and Optimization Issues,in Advances in Neural Information Processing Systems 3, R.L. Lippmann, J.E. Moody, D.S. Touretzky (eds.) 1989Google Scholar
  49. [49]
    Rumelhart, D.E., Hinton, G.E., Williams, R.J. Learning Internal Representation by Error Propagation, in: Rumelhart, D.E., McClelland, J.L., and the PDP Research Group, Parallel Distributed Processing, Explorations in the Microstructure of Cognition, Vol. 1: Foundations, MIT Press, Cambridge, Massachusetts, S. 318 ff. (1986)Google Scholar
  50. [50]
    Rumelhart, D.E., McClelland, J.L. (eds.), Parallel Distributed Processing, Vol. 1,2, MIT Press, Cambridge, Massachusetts (1986)Google Scholar
  51. [51]
    Sankar, Anath and Mammone, Richard, J., Growing and Pruning Neural Tree Networks, IEEE Transactions on Computers, Vol. 42, NO. 3, (1993)Google Scholar
  52. [52]
    Schepers, A., Arrninger, G. und Küsters, U., The Analysis of Non-Metric Endogenous Variables in Latent Variable Models: The Mecosa Approach, in P. Gruber (ed.), Econometric Decision Models: New Methods of Modeling and Applications, Springer Verlag, Heidelberg, 459–472 (1991)Google Scholar
  53. [53]
    Silva, F. M., Almeida, L.B., Speeding up Backpropagation, in: Advanced Neural Computers, R. Eckmiller (ed.) (1990)Google Scholar
  54. [54]
    Stock, J.H., Asymptotic Properties of Least Squares Estimators of Cointe-grating Vectors, Econometrics 55, 1035–1056 (1987)CrossRefGoogle Scholar
  55. [55]
    Straub, R., Schwarz, D., Schöneburg, E. Simulation of Backpropagation Networks on Transputers, Suitable Topologies and Performance Analysis, in: Neurocomputing, Vol. 2, S. 199 ff. (1991)Google Scholar
  56. [56]
    Weigend, A.S., Huberman, B.A., Rumelhart, D.E. Predicting the Future: A Connectionist Approach, Stanford University, Stanford, California (1990)Google Scholar
  57. [57]
    White, H., Learning in Artificial Neural Networks: A Statistical Perspective, Neural Computation, 1, 425–464 (1989a)CrossRefGoogle Scholar
  58. [58]
    White, H., Some Asymptotic Results of Learning in Single Hidden-Layer Feedforward Network Models, Journal of the American Statistical Association, Vol. 84, No. 408, 1003–1013 (1989b)Google Scholar
  59. [59]
    Whitley, D., Dominic, S. and Das R., Genetic Reinforcement Learning with Multilayer Neural Networks, in: Proc. Fourth Intern. Conf. Genetic Algorithms, Morgan Kaufmann, San Mateo, CA (1991)Google Scholar
  60. [60]
    Zimmermann, H.G., Hergert, F., Finnoff, W. Neuron Pruning and Merging Methods for Use in Conjunction with Weight Elimination, Siemens AG, Corporate Research and Development, ZFE IS INF 23, Otto-Hahn-Ring 6, 8000 München 83 (1992)Google Scholar

Copyright information

© Physica-Verlag Heidelberg 1993

Authors and Affiliations

  • Walter Frisch
    • 1
  1. 1.Institut für Informationsverarbeitung und Informations Wirtschaft, Abteilung für Angewandte Informatik insbesondere BetriebsinformatikWirtschaftsuniversität WienWienÖsterreich

Personalised recommendations