Abstract
The dynamics of a two neuron artificial recurrent neural network depends on six parameters: four synaptic weights and two external inputs. There are complicated relationships between these parameters and the behavior of the system. The full parameter space has not been studied yet and has been limited to detect behaviors for specific configurations, i.e., when some parameters are pre-set, either for two or three neurons. In this study we analyze the nature of the fixed point at the origin in a two-neuron discrete recurrent neural network by plotting the bifurcation manifolds in the full weights space which is a 4-dimensinal one that gives a clear view of what the dynamics of the system can be around this fixed point, which is a very influent point in the global dynamics of the system. The possible bifurcations at the origin are Saddle, Period Doubling and Neimark-Sacker. We found, among other results, that the Neimark-Sacker bifurcation is only possible when the synaptic connections between neurons are one excitatory and one inhibitory.
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
Beer, R.D.: Parameter Space Structure of Continuous-Time Recurrent Neural Networks. Neural Computation 18(12), 3009–3051 (2006)
Cervantes, J., Gómez, M., Schaum, A.: Some Implications of System Dynamics Analysis of Discrete-Time Recurrent Neural Networks for Learning Algorithms Design. In: 12th Mexican International Conference on Artificial Intelligence, pp. 73–79 (2013)
Hajihosseini, A., Maleki, F., Rokni Lamooki, G.R.: Bifurcation analysis on a generalized recurrent neural network with two interconnected three-neuron components. Chaos, Solitons & Fractals 44(11), 1004–1019 (2011)
Haschke, R., Steil, J.J.: Input space bifurcation manifolds of recurrent neural networks. Neurocomputing 64, 25–38 (2005)
He, W., Cao, J.: Stability and bifurcation of a class of discrete-time neural networks. Applied Mathematical Modelling 31, 2111–2122 (2007)
Kuznetsov, Y.A.: Elements of applied bifurcation theory (2004)
Pasemann, F.: Complex dynamics and the structure of small neural networks. Network: Computation in Neural Systems 13(2), 195–216 (2002)
Author information
Authors and Affiliations
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2014 Springer International Publishing Switzerland
About this paper
Cite this paper
Cervantes-Ojeda, J., del Carmen Gómez-Fuentes, M. (2014). On the Connection Weight Space Structure of a Two-Neuron Discrete Neural Network: Bifurcations of the Fixed Point at the Origin. In: Gelbukh, A., Espinoza, F.C., Galicia-Haro, S.N. (eds) Nature-Inspired Computation and Machine Learning. MICAI 2014. Lecture Notes in Computer Science(), vol 8857. Springer, Cham. https://doi.org/10.1007/978-3-319-13650-9_8
Download citation
DOI: https://doi.org/10.1007/978-3-319-13650-9_8
Publisher Name: Springer, Cham
Print ISBN: 978-3-319-13649-3
Online ISBN: 978-3-319-13650-9
eBook Packages: Computer ScienceComputer Science (R0)