Abstract
Recurrent neural networks (RNNs) are typically considered as relatively simple architectures, which come along with complicated learning algorithms. This paper has a different view: We start from the fact that RNNs can model any high dimensional, nonlinear dynamical system. Rather than focusing on learning algorithms, we concentrate on the design of network architectures. Unfolding in time is a well-known example of this modeling philosophy. Here a temporal algorithm is transferred into an architectural framework such that the learning can be performed by an extension of standard error backpropagation.
We introduce 12 tricks that not only provide deeper insights in the functioning of RNNs but also improve the identification of underlying dynamical system from data.
This is a preview of subscription content, log in via an institution to check access.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Preview
Unable to display preview. Download preview PDF.
References
Calvert, D., Kremer, S.: Networks with Adaptive State Transitions. In: Kolen, J.F., Kremer, S. (eds.) A Field Guide to Dynamical Recurrent Networks, pp. 15–25. IEEE (2001)
Föllmer, H.: Alles richtig und trotzdem falsch?, Anmerkungen zur Finanzkrise und Finanzmathematik. In: MDMV, vol. 17, pp. 148–154 (2009)
Haykin, S.: Neural Networks and Learning Machines, 3rd edn. Prentice Hall (2008)
Hull, J.: Options, Futures & Other Derivative Securities. Prentice Hall (2001)
Hornik, K., Stinchcombe, M., White, H.: Multilayer Feedforward Networks are Universal Approximators. Neural Networks 2, 359–366 (1989)
Kamien, M., Schwartz, N.: Dynamic Optimization: The Calculus of Variations and Optimal Control in Economics and Management, 2nd edn. Elsevier Science (October 1991)
McNeil, A., Frey, R., Embrechts, P.: Quantitative Risk Management: Concepts, Techniques and Tools. New Jersey. Princeton University Press, Princeton (2005)
Neuneier, R., Zimmermann, H.-G.: How to Train Neural Networks. In: Orr, G.B., Müller, K.-R. (eds.) NIPS-WS 1996. LNCS, vol. 1524, pp. 373–423. Springer, Heidelberg (1998)
Pearlmutter, B.: Gradient Calculations for Dynamic Recurrent Neural Networks. In: Kolen, J.F., Kremer, S. (eds.) A Field Guide to Dynamical Recurrent Networks, pp. 179–206. IEEE Press (2001)
Rumelhart, D.E., Hinton, G.E., Williams, R.J.: Learning Internal Representations by Error Propagation. In: Rumelhart, D.E., McClelland, J.L., et al. (eds.) Parallel Distributed Processing, vol. 1. MIT Press, Cambridge (1986)
Schäfer, A.M., Zimmermann, H.-G.: Recurrent Neural Networks Are Universal Approximators. In: Kollias, S.D., Stafylopatis, A., Duch, W., Oja, E. (eds.) ICANN 2006. LNCS, vol. 4131, pp. 632–640. Springer, Heidelberg (2006)
Wei, W.S.: Time Series Analysis: Univariate and Multivariate Methods. Addison-Wesley Publishing Company, N.Y. (1990)
Werbos, P.J.: Beyond Regression: New Tools for Prediction and Analysis in the Behavioral Sciences. PhD Thesis, Harvard University (1974)
Williams, R.J., Zipser, D.: A Learning Algorithm for continually running fully recurrent neural networks. Neural Computation 1(2), 270–280 (1989)
Zimmermann, H.G., Grothmann, R., Neuneier, R.: Modeling of Dynamical Systems by Error Correction Neural Networks. In: Soofi, A., Cao, L. (eds.) Modeling and Forecasting Financial Data, Techniques of Nonlinear Dynamics. Kluwer (2002)
Zimmermann, H.G., Grothmann, R., Schäfer, A.M., Tietz, Ch.: Modeling Large Dynamical Systems with Dynamical Consistent Neural Networks. In: Haykin, S., et al. (eds.) New Directions in Statistical Signal Processing. MIT Press, Cambridge (2006)
Zimmermann, H.G.: Neuronale Netze als Entscheidungskalkül. In: Rehkugler, H., Zimmermann, H.G. (eds.) Neuronale Netze in der Ökonomie, Grundlagen und wissenschaftliche Anwendungen, Vahlen, Munich (1994)
Zimmermann, H.G., Neuneier, R.: Neural Network Architectures for the Modeling of Dynamical Systems. In: Kolen, J.F., Kremer, S. (eds.) A Field Guide to Dynamical Recurrent Networks, pp. 311–350. IEEE Press (2001)
Zimmermann, H.G., Grothmann, R., Tietz, C., von Jouanne-Diedrich, H.: Market Modeling, Forecasting and Risk Analysis with Historical Consistent Neural Networks. In: Hu, B., et al. (eds.) Proceedings of Operations Research 2010, Selected Papers of the Annual Int. Conferences of the German OR Society (GOR), Munich. Springer, Heidelberg (2011)
Zimmermann, H.G., Grothmann, R., Tietz, C.: Forecasting Market Prices with Causal-Retro-Causal Neural Networks. In: Klatte, D., [ERROR while converting LaTeX/Unicode], H.-J., Schmedders, K. (eds.) Proceedings of Operations Research 2011, Selected Papers of the Int. Conference on Operations Research 2011 (OR 2011), Zurich, Switzerland. Springer (2012)
Author information
Authors and Affiliations
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2012 Springer-Verlag Berlin Heidelberg
About this chapter
Cite this chapter
Zimmermann, HG., Tietz, C., Grothmann, R. (2012). Forecasting with Recurrent Neural Networks: 12 Tricks. In: Montavon, G., Orr, G.B., Müller, KR. (eds) Neural Networks: Tricks of the Trade. Lecture Notes in Computer Science, vol 7700. Springer, Berlin, Heidelberg. https://doi.org/10.1007/978-3-642-35289-8_37
Download citation
DOI: https://doi.org/10.1007/978-3-642-35289-8_37
Publisher Name: Springer, Berlin, Heidelberg
Print ISBN: 978-3-642-35288-1
Online ISBN: 978-3-642-35289-8
eBook Packages: Computer ScienceComputer Science (R0)