Submit your paper : editorIJETjournal@gmail.com

Paper Title : A Comprehensive Comparison of Artificial and Spiking Neural Networks

ISSN : 2395-1303
Year of Publication : 2021
10.29126/23951303/IJET-V7I4P04

Authors: Vinanth S Bharadwaj,Anupama P

         


Abstract
- In this paper, we discuss available artificial neural networks and Spiking Neural Networks and their ways of solving the problem of Handwritten character recognition by using the MNIST dataset which is an open-source dataset.MNIST is a subset of NSIT dataset and has a set of over 60000 images of handwritten numbers. There is a pursuit to build something similar to the human brain. The human brain is the best computer there isand very energy efficient too. Attempts and various frameworks have been introduced to mimic the human brain but there is still a worthy mile to be covered before we build a computer as wonderful as the human brain. We compare the Convolution Neural Network, Self Organizing Maps, and Supervised and Unsupervised Spiking neural networks which classify the MNIST dataset and also give an overview of their efficiencies.

Reference
[1] Filipp Akopyan. “Design and Tool Flow of IBM’s TrueNorth: An Ultra-Low Power Programmable Neu rosynaptic Chip with 1 Million Neurons”. In: Pro ceedings of the 2016 on International Symposium on Physical Design. ISPD ’16. Santa Rosa, California, USA: Association for Computing Machinery, 2016, pp. 59–60. ISBN: 9781450340397. DOI: 10 . 1145 / 2872334 . 2878629. URL: https : / / doi . org / 10 . 1145 / 2872334.2878629. [2] Li Deng. “The MNIST Database of Handwritten Digit Images for Machine Learning Research [Best of the Web]”. In: IEEE Signal Processing Magazine 29.6 (2012), pp. 141–142. DOI: 10.1109/MSP.2012.2211477. [3] Peter Diehl and Matthew Cook. “Unsupervised learning of digit recognition using spike-timingdependent plas ticity”. In: Frontiers in Computational Neuroscience 9 (2015), p. 99. ISSN: 1662-5188. DOI: 10.3389/fncom. 2015.00099. URL: https://www.frontiersin.org/article/ 10.3389/fncom.2015.00099. [4] Wulfram Gerstner and Werner M. Kistler. Spiking Neu ron Models: Single Neurons, Populations, Plasticity. Cambridge University Press, 2002. DOI: 10 . 1017 / CBO9780511815706. [5] SAMANWOY GHOSH-DASTIDAR and HOJJAT ADELI. “SPIKING NEURAL NETWORKS”. In: International Journal of Neural Systems 19.04 (2009). PMID: 19731402, pp. 295–308. DOI: 10 . 1142 / S0129065709002002. eprint: https : / / doi . org / 10 . 1142 / S0129065709002002. URL: https://doi.org/10.1142/S0129065709002002. [6] Hananel Hazan et al. “BindsNET: A Machine Learning Oriented Spiking Neural Networks Library in Python”. In: Frontiers in Neuroinformatics 12 (2018), p. 89. ISSN: 1662- 5196. DOI: 10.3389/fninf.2018.00089. URL: https://www.frontiersin.org/article/10.3389/fninf.20 18. 00089. [7] T. Kohonen. “The self-organizing map”. In: Proceed ings of the IEEE 78.9 (1990), pp. 1464–1480. DOI: 10.1109/5.58325. [8] Y. LeCun et al. “Gradient-Based Learning Applied to Document Recognition”. In: Intelligent Signal Process ing. IEEE Press, 2001, pp. 306–351. [9] Martın Abadi et al. TensorFlow: Large-Scale Machine Learning on Heterogeneous Systems. Software avail able from tensorflow.org. 2015. URL: https : / / www . tensorflow.org/. [10] Jianguo Xin and M.J. Embrechts. “Supervised learning with spiking neural networks”. In: IJCNN’01. Interna tional Joint Conference on Neural Networks. Proceed ings (Cat. No.01CH37222). Vol. 3. 2001, 1772–1777 vol.3. DOI: 10.1109/IJCNN.2001.938430
Keywords
——- Convolution Neural Networks, Spiking Neural Networks, Hebbian learning, Spiking Neural Networks.