Abstract
It is known that the problem of training certain specific, very simple neural networks is NP-complete. While such results suggest that training is equally hard for larger, as well as differently configured networks, this conclusion is by no means self-evident. The main result of this paper is that it is NP-complete to train any specific architecture or class of architectures. On the other hand, it is also shown that a simple 4-node network (with two hidden and two output units) can be trained in polynomial time if the target network function is assumed to be surjective. This remains true for networks of any size, on the condition that the number of ouput units is at least equal to the number of units in the first computing layer. Thus, with a mild restriction placed on the class of target functions, training certain large networks may be easier than training smaller ones.
Preview
Unable to display preview. Download preview PDF.
References
Baum, E.B.: On learning a union of halfspaces. Journal of Complexity 6 (1990) 67–101
Baum, E.B.: A polynomial time algorithm that learns two hidden unit nets. Neural Computation 2 (1990) 510–522
Baum, E.B.: Neural net algorithms that learn in polynomial time from examples and queries. IEEE Trans. on Neural Networks 2 (1991) 5–19
Blum, A.L., Rivest, R.L.: Training a 3-node neural network is NP-complete. Machine Learning: From Theory to Applications, Springer-Verlag (1993) 9–28
DasGupta, B., Siegelmann, H., Sontag, E.: On a learnability question associated with neural networks with continuous activations. Proc. of the 7th Annual ACM Conference on Computational Learning Theory (1994) 47–56
Golea, M., Hancock, T., Marchand, M.: Learning nonoverlapping perceptron networks from examples and membership queries. Machine Learning 16 (1994) 161–183
Judd, J.S.: On the complexity of loading shallow neural networks. Journal of Complexity 4 (1988) 177–192
Lin, J.H., Vitter, J.S.: Complexity results on learning by neural nets. Machine Learning 6 (1991) 211–230
Author information
Authors and Affiliations
Editor information
Rights and permissions
Copyright information
© 1995 Springer-Verlag Berlin Heidelberg
About this paper
Cite this paper
Pinter, C.C. (1995). Complexity of network training for classes of Neural Networks. In: Jantke, K.P., Shinohara, T., Zeugmann, T. (eds) Algorithmic Learning Theory. ALT 1995. Lecture Notes in Computer Science, vol 997. Springer, Berlin, Heidelberg. https://doi.org/10.1007/3-540-60454-5_40
Download citation
DOI: https://doi.org/10.1007/3-540-60454-5_40
Published:
Publisher Name: Springer, Berlin, Heidelberg
Print ISBN: 978-3-540-60454-9
Online ISBN: 978-3-540-47470-8
eBook Packages: Springer Book Archive