Abstract
For the last few decades, the neuroscientific research has highlighted the importance of astrocytes, a type of glial cells, in the information processing capabilities. By dynamic bidirectional communication with neurons, astrocytes regulate their excitability through a variety of mechanisms. Traditional artificial neural networks (ANNs) are connectionist models that describe how information passes throughout layer of neurons abstracting from low-level mechanisms. However, very little research has addressed artificial astrocytes and their incorporation into ANNs. In this paper, we present an echo state network (ESN) extended with astrocytes which influence the neurons by fixed or Hebbian connections. By systematic analysis we investigate their role on five classification tasks and show that they can outperform the standard ESN without astrocytes. Although the model with fixed astrocytic weights yields from none to little improvement, the model with Hebbian weights from astrocytes to neurons is significantly superior.
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
Notes
- 1.
Originally, authors use term artificial glia but we consider artificial astrocytes instead, since glia represent the vast majority of non-neuronal cells in the nervous system with multiple functions, whereas only astrocytes are currently considered to play a vital role in information processing tasks.
References
Alvarellos-González, A., Pazos, A., Porto-Pazos, A.B.: Computational models of neuron-astrocyte interactions lead to improved efficacy in the performance of neural networks. Comput. Math. Methods Med. 2012, 10 pages (2012)
Alvarez-Maubecin, V., García-Hernández, F., Williams, J.T., Van Bockstaele, E.J.: Functional coupling between neurons and GLIA. J. Neurosci. 20(11), 4091–4098 (2000)
Azevedo, F.A., et al.: Equal numbers of neuronal and nonneuronal cells make the human brain an isometrically scaled-up primate brain. J. Comp. Neurol. 513(5), 532–541 (2009)
Chen, Y., et al.: The UCR time series classification archive (2015). www.cs.ucr.edu/~eamonn/time_series_data/
Dallérac, G., Chever, O., Rouach, N.: How do astrocytes shape synaptic transmission? insights from electrophysiology. Front. Cell. Neurosci. 7, 159 (2013)
Fellin, T., Pascual, O., Haydon, P.G.: Astrocytes coordinate synaptic networks: balanced excitation and inhibition. Physiology 21(3), 208–215 (2006)
Gergel’, P., Farkaŝ, I.: Investigating the role of astrocyte units in a feedforward neural network. In: Kurková, V., Manolopoulos, Y., Hammer, B., Iliadis, L., Maglogiannis, I. (eds.) ICANN 2018. LNCS, vol. 11141, pp. 73–83. Springer, Cham (2018). https://doi.org/10.1007/978-3-030-01424-7_8
Haydon, P.G.: Neuroglial networks: neurons and glia talk to each other. Curr. Biol. 10(19), R712–R714 (2000)
Hebb, D.O.: The Organization of Behavior: A Neuropsychological Theory. Wiley, New York (1949)
Ikuta, C., Uwate, Y., Nishio, Y.: Chaos glial network connected to multi-layer perceptron for solving two-spiral problem. In: Proceedings of 2010 IEEE International Symposium on Circuits and Systems, pp. 1360–1363 (2010)
Ikuta, C., Uwate, Y., Nishio, Y.: Multi-layer perceptron with impulse glial network. In: IEEE Workshop on Nonlinear Circuit Networks, pp. 9–11 (2010)
Ikuta, C., Uwate, Y., Nishio, Y.: Performance and features of multi-layer perceptron with impulse glial network. In: International Joint Conference on Neural Networks, pp. 2536–2541 (2011)
Ikuta, C., Uwate, Y., Nishio, Y.: Multi-layer perceptron with positive and negative pulse glial chain for solving two-spirals problem. In: International Joint Conference on Neural Networks, pp. 1–6 (2012)
Jaeger, H.: The “echo state" approach to analysing and training recurrent neural networks-with an erratum note. Bonn, Ger.: Ger. Nat. Res. Cent. Inf. Technol. GMD Tech. Rep. 148(34), 13 (2001)
Matthews, B.W.: Comparison of the predicted and observed secondary structure of t4 phage lysozyme. Biochimica et Biophysica Acta (BBA)-Protein Struct. 405(2), 442–451 (1975)
Mesejo, P., Ibánez, O., Fernández-Blanco, E., Cedrón, F., Pazos, A., Porto-Pazos, A.B.: Artificial neuron-glia networks learning approach based on cooperative coevolution. Int. J. Neural Syst. 25(04), 1550012 (2015)
Oja, E.: Simplified neuron model as a principal component analyzer. J. Math. Biol. 15(3), 267–273 (1982)
Oschmann, F., Berry, H., Obermayer, K., Lenk, K.: From in silico astrocyte cell models to neuron-astrocyte network models: a review. Brain Res. Bull. 136, 76–84 (2018)
Porter, J.T., McCarthy, K.D.: Astrocytic neurotransmitter receptors in situ and in vivo. Prog. Neurobiol. 51(4), 439–455 (1997)
Porto-Pazos, A.B., et al.: Artificial astrocytes improve neural network performance. PloS ONE 6(4), e19109 (2011)
Volman, V., Bazhenov, M., Sejnowski, T.J.: Computational models of neuron-astrocyte interaction in epilepsy. Front. Comput. Neurosci. 6, 58 (2012)
Wade, J., Kelso, S., Crunelli, V., McDaid, L.J., Harkin, J.: Biophysically based computational models of astrocyte-neuron coupling and their functional significance. Frontiers E-books (2014)
Acknowledgments
This work was supported by grant UK/250/2019 from Comenius University in Bratislava (P.G.) and Slovak Grant Agency for Science, project VEGA 1/0796/18 (I.F.).
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2019 Springer Nature Switzerland AG
About this paper
Cite this paper
Gergel’, P., Farkaš, I. (2019). Echo State Networks with Artificial Astrocytes and Hebbian Connections. In: Rojas, I., Joya, G., Catala, A. (eds) Advances in Computational Intelligence. IWANN 2019. Lecture Notes in Computer Science(), vol 11506. Springer, Cham. https://doi.org/10.1007/978-3-030-20521-8_38
Download citation
DOI: https://doi.org/10.1007/978-3-030-20521-8_38
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-030-20520-1
Online ISBN: 978-3-030-20521-8
eBook Packages: Computer ScienceComputer Science (R0)