Solving Number Series with Simple Recurrent Networks
Number series tests are a popular task in intelligence tests to measure a person’s ability of numerical reasoning. The function represented by a number series can be learned by artificial neural networks. In contrast to earlier research based on feedforward networks, we apply simple recurrent networks to the task of number series prediction. We systematically vary the number of input and hidden units in the networks to determine the optimal network configuration for the task. While feedforward networks could solve only 18 of 20 test series, a very small simple recurrent network could find a solution for all series. This underlines the importance of recurrence in such systems, which further is a basic concept in human cognition.
KeywordsHide Layer Number Series Recurrent Neural Network Hide Unit Implicit Learning
Unable to display preview. Download preview PDF.
- 3.Bishop, C.M.: Neural Networks for Pattern Recognition. Oxford University Press, Inc., New York (1995)Google Scholar
- 4.Cleeremans, A.: Mechanisms of Implicit Learning: Connectionist Models of Sequence Processing. Neural Network Modeling & Connectionism. Mit Press (1993)Google Scholar
- 6.Erhan, D., Pierre-Antoine, M., Bengio, Y., Bengio, S., Vincent, P.: The difficulty of training deep architectures and the effect of unsupervised pre-training. In: Proceedings of the 12th International Conference on Artificiall Intelligence and Statistics, pp. 153–160 (2009)Google Scholar
- 7.Glüge, S., Böck, R., Wendemuth, A.: Implicit sequence learning - a case study with a 4-2-4 encoder simple recurrent network. In: IJCCI (ICFC-ICNC), pp. 279–288 (2010)Google Scholar
- 12.Park, D.C.: Sunspot series prediction using adaptively trained multiscale-bilinear recurrent neural network. In: 9th IEEE International Conference on Computer Systems and Applications (AICCSA), pp. 135–139 (2011)Google Scholar
- 14.Ragni, M., Klein, A.: Solving number series - architectural properties of successful artificial neural networks. In: Madani, K., Kacprzyk, J., Filipe, J. (eds.) IJCCI (NCTA), pp. 224–229. SciTePress (2011)Google Scholar