Abstract
Gradient descent is one of the significant research contents in supervised learning of spiking neural networks (SNNs). In order to improve the performance of gradient descent learning algorithms for multilayer SNNs, this paper proposes a spike selection mechanism to select optimal presynaptic spikes to participate in computing the change amount of synaptic weights during the process of weight adjustment. The proposed spike selection mechanism comprehensively considers the desired and actual output spikes of the network. The presynaptic spikes involved in the calculation are determined within a certain time interval, so that the network output spikes matches the desired output spikes perfectly as far as possible. The proposed spike selection mechanism is used for the gradient descent learning algorithm for multilayer SNNs. The experimental results show that our proposed mechanism can make the gradient descent learning algorithm for multilayer SNNs have higher learning accuracy, fewer learning epochs and shorten the running time. It indicates that the spike selection mechanism is very effective for improving gradient descent learning performance.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
References
Ghosh-Dastidar, S., Adeli, H.: Spiking neural networks. Int. J. Neural Syst. 19(04), 295–308 (2009)
Taherkhani, A., Belatreche, A., Li, Y., et al.: A review of learning in biologically plausible spiking neural networks. Neural Netw. 122, 253–272 (2020)
Skatchkovsky, N., Jang, H., Simeone, O.: Spiking neural networks-part II: detecting spatio-temporal patterns. IEEE Commun. Lett. 25(6), 1741–1745 (2021)
Kulkarni, S.R., Rajendran, B.: Spiking neural networks for handwritten digit recognition-supervised learning and network optimization. Neural Netw. 103, 118–127 (2018)
Wang, X., Lin, X., Dang, X.: Supervised learning in spiking neural networks: a review of algorithms and evaluations. Neural Netw. 125, 258–280 (2020)
Lin, X., Wang, X., Zhang, N., et al.: Supervised learning algorithms for spiking neural networks: a review. Acta Electron. Sin. 43(3), 577–586 (2015)
Lin, X., Wang, X.: Spiking Neural Networks: Principles and Applications. Science Press, China (2018)
Comsa, I., Potempa, K., Versari, L., et al.: Temporal coding in spiking neural networks with alpha synaptic function: learning with backpropagation. IEEE Trans. Neural Netw. Learn. Syst., 1–14 (2021)
Kheradpisheh, S., Masquelier, T.: Temporal backpropagation for spiking neural networks with one spike per neuron. Int. J. Neural Syst. 30(06), 2050027 (2020)
Bohte, S.M., Kok, J.N., Poutré, H.: Error-backpropagation in temporally encoded networks of spiking neurons. Neurocomputing 48(1), 17–37 (2002)
Zhao, J., Zurada, J.M., Yang, J., Wu, W.: The convergence analysis of SpikeProp algorithm with smoothing \({L_{1/2}}\) regularization. Neural Netw. 103, 19–28 (2018)
Shrestha, S.B., Song, Q.: Robustness to training disturbances in SpikeProp learning. IEEE Trans. Neural Netw. Learn. Syst. 29(7), 3126–3139 (2018)
Booij, O., tat Nguyen, H.: A gradient descent rule for spiking neurons emitting multiple spikes. Inf. Process. Lett. 95(6), 552–558 (2005)
Xu, Y., Zeng, X., Han, L., Yang, J.: A supervised multi-spike learning algorithm based on gradient descent for spiking neural networks. Neural Netw. 43, 99–113 (2013)
Xu, Y., Yang, J., Zhong, S.: An online supervised learning method based on gradient descent for spiking neurons. Neural Netw. 93, 7–20 (2017)
Gerstner, W., Kistler, W.M.: Spiking Neuron Models: Single Neurons, Populations, Plasticity. Cambridge University Press, Cambridgeshire (2002)
Acknowledgments
This work was supported by the National Natural Science Foundation of China under grant no. 61762080, the Key Research and Development Project of Gansu Province under grant no. 20YF8GA049, the Youth Science and Technology Fund Project of Gansu Province under grant no. 20JR10RA097, the Lanzhou Municipal Science and Technology Project under grant no. 2019-1-34.
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2021 Springer Nature Switzerland AG
About this paper
Cite this paper
Lin, X., Hu, T., Wang, X., Lu, H. (2021). Gradient Descent Learning Algorithm Based on Spike Selection Mechanism for Multilayer Spiking Neural Networks. In: Mantoro, T., Lee, M., Ayu, M.A., Wong, K.W., Hidayanto, A.N. (eds) Neural Information Processing. ICONIP 2021. Lecture Notes in Computer Science(), vol 13110. Springer, Cham. https://doi.org/10.1007/978-3-030-92238-2_4
Download citation
DOI: https://doi.org/10.1007/978-3-030-92238-2_4
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-030-92237-5
Online ISBN: 978-3-030-92238-2
eBook Packages: Computer ScienceComputer Science (R0)