On the approximation capability of recurrent neural networks

Neurocomputing

Volumn 31, Issue 1-4, 2000, Pages 107-123

  Hammer, Barbara ^a  

^a UNIVERSITY OF OSNABRÜCK   (Germany)

Author keywords

Computational capability; Recurrent neural networks; Sigmoidal networks; Universal approximation

Indexed keywords

APPROXIMATION THEORY; FUNCTIONS; PROBABILITY; VECTORS;

SIGMOIDAL NETWORKS;

RECURRENT NEURAL NETWORKS;

ALGORITHM; ARTIFICIAL NEURAL NETWORK; LINEAR SYSTEM; MATHEMATICAL COMPUTING; PRIORITY JOURNAL; PROBABILITY; REVIEW;

EID: 0034069365     PISSN: 09252312     EISSN: None     Source Type: Journal    
DOI: 10.1016/S0925-2312(99)00174-5     Document Type: Article

References (21)

1. P. Alexandroff, H. Hopf, Topologie, Vol. 1, Springer, Berlin, 1974.
2. A. Eliseeff, H. Paugam-Moisy, Size of multilayer networks for exact learning: analytic approach, in: M.C. Mozer, M.I. Jordan, T. Petsche (Eds.), Advances in Neural Information Processing Systems, Vol. 9, The MIT Press, Cambridge, MA, 1997, pp. 162-168.
3. Elman J.L. Distributed representations, simple recurrent networks, and grammatical structure. Mach. Learning. 7:1991;195-225.
4. C.L. Giles, M. Gori (Eds.), Adaptive Processing of Sequences and Data Structures, Springer, Berlin, 1998.
5. C. Goller, A. Küchler, Learning task-dependent distributed representations by backpropagation through structure, in: Proceedings of the IEEE International Conference on Neural Networks, 1996, pp. 347-352.
6. Hammer B. On the learnability of recursive data. Math. Control Signals Systems. 12:1999;62-79.
7. B. Hammer, V. Sperschneider, Neural networks can approximate mappings on structured objects, in: P.P. Wang (Ed.), International Conference of Information Sciences, Vol. 2, 1997, pp. 211-214.
8. Hochreiter S., Schmidhuber J. Long short-term memory. Neural Comput. 9:1997;1735-1780.
9. Hornik K. Some new results on neural network approximation. Neural Networks. 6:1993;1069-1072.
10. Hornik K., Stinchcombe M., White H. Multilayer feedforward networks are universal approximators. Neural Networks. 2:1989;359-366.
11. Kilian J., Siegelmann H.T. The dynamic universality of sigmoidal neural networks. Inform. and Comput. 128:1996;48-56.
12. Lippmann R. Review of neural networks for speech recognition. Neural Comput. 1:1989;1-38.
13. Maass W., Orponen P. On the effect of analog noise in discrete-time analog computation. Neural Comput. 10:1998;1071-1095.
14. M. Mozer, Neural net architectures for temporal sequence processing, in: A. Weigend, N. Gershenfeld (Eds.), Predicting the Future and Understanding the Past, Addison-Wesley, Reading, MA, 1993.
15. Narendra K.S., Parthasarathy K. Identification and control of dynamical systems using neural networks. IEEE Trans. Neural Networks. 1:1990;4-27.
16. S. Schulz, A. Küchler, C. Goller, Some experiments on the applicability of folding architectures to guide theorem proving, in: D.D. Dankel (Ed.), Proceedings of the 10th International FLAIRS Conference, AI Research Society, Florida, 1997, pp. 377-381.
17. Siegelmann H.T., Sontag E.D. Analog computation, neural networks, and circuits. Theoret. Comput. Sci. 131:1994;331-360.
18. Siegelmann H.T., Sontag E.D. On the computational power of neural neworks. J. Comput. System Sci. 50:1995;132-150.
19. Sontag E.D. Feedforward nets for interpolation and classification. J. Comput. System Sci. 45:1992;20-48.
20. E.D. Sontag, Neural nets as systems models and controllers, in: Seventh Yale Workshop on Adaptive and Learning Systems, 1992, pp. 73-79.
21. Williams R., Zipser D. A learning algorithm for continually running fully recurrent neural networks. Neural Comput. 1:1989;270-280.