Indian Journal of Science and Technology

Open Access Open Access  Restricted Access Subscription Access

Massively Parallel Computational Schemes for Simulating Spiking Neural Networks using GPU Accelerators


Affiliations
1 Department of Computer Science and Engineering, Jain University, Nitte Meenakshi Institute of Technology, P.O. Box 6429, Yelahanka, Bengaluru – 560064, Karnataka, India
2 Centre for Incubation, Innovation, Research and Consultancy, Jyothy Institute of Technology, Tataguni, Off Kanakapura Road, Bengaluru – 560082, Karnataka, India
 

Objectives: To review various tools available for simulating Spiking Neural Networks using heterogeneous parallel processing platforms that help to reduce cost, increase the computational speed and also to document/archive lessons learnt. Methods/Statistical Analysis: The computational speed is a continuing challenge for simulating genuine spiking neural network models. Understanding of the spiking neural networks is significantly simplified by computer simulators like NEST, GeNN, EDLUT and BRIAN. Findings: Simulation is a handy toolkit of scientists and engineers of all disciplines. NEST, GeNN, EDLUT and BRIAN simulators help in achieving better performance not in terms of same kind of processing but with additional special tasks which require more computational power. BRIAN and EDLUT which are hybrid simulators supports both time driven and event driven techniques and outperform when compared to other simulators. Application/Improvements: Using BRIAN and EDLUT simulation techniques we can achieve the high performance when compared to other spiking neural simulation techniques.
User

  • Gerstner W, Kistler W. Spiking Neuron Models. Cambridge University Press; 2002. p. 1–28. https://doi.org/10.1017/CBO9780511815706, https://doi.org/10.1017/CBO9780511815706.002. PMid: 11842739.

  • Hwu WM. GPU Computing Gems Emerald Edition, 1st Edition. Morgan Kaufman; 2011.

  • Carnevale N, Hines M. The NEURON Book. Cambridge University Press; 2006. https://doi.org/10.1017/CBO9780511541612.

  • Izhikevich EM. Dynamical Systems in Neuroscience: The Geometry of Excitability and Bursting. The MIT Press Cambridge; 2007. PMid: 17220510, PMCid: PMC4437488.

  • Vreeken J. Spiking neural networks: An introduction, Tech. Rep., Artificial Intelligence laboratory, Intelligent Systems Group, University of Utrecht, 2003.

  • Brette R. Simulation of networks of spiking neurons: A review of tools and strategies, Journal of Computational Neuroscience. 2007; 23(3):349–98. https://doi.org/10.1007/s10827-007-0038-6. PMid: 17629781, PMCid: PMC2638500.

  • Brette R, Goodman DF. Simulating spiking neural networks on GPU, Network. 2012; 23(4):167–82. https://doi.org/10.3109/0954898X.2012.730170. PMid: 23067314.

  • Chessa M, Bianchi V, Zampetti M, Sabatini SP, Solari F. Real-time simulation of large-scale neural architectures for visual features computation based on GPU; Network. 2012; 23(4):272–91. https://doi.org/10.3109/09548 98X.2012.737500. PMid: 23116085.

  • Anantha Narayanan R, Modha DS. Anatomy of a cortical simulator. SC '07: Proceedings of the 2007 ACM/IEEE Conference on Super Computing; 2007. p. 1–12. https://doi.org/10.1145/1362622.1362627.

  • NEST Simulator. Date accessed: 2016. http://www.nestinitiative.org/.

  • Carlson KD, Nageswaran JM, Dutt N, Krichmar JL. An efficient automated parameter tuning framework for spiking neural networks, Neuroscience. 2014; 8(10):1–16.

  • Fountas Z, Shanahan M. GPU-based fast parameter optimization for phenomenological spiking neural models. International Joint Conference on Neural Networks (IJCNN), Killarney, Ireland; 2015. p. 1–8. https://doi.org/10.1109/IJCNN.2015.7280668.

  • Izhikevich EM. Which model to use for cortical spiking neurons? IEEE Transaction Neural Networking. 2004; 15(5):1063–70. https://doi.org/10.1109/TNN.2004.832719. PMid: 15484883.

  • Hodgkin AL, Huxley AF. A quantitative description of membrane current and its application to conduction and excitation in nerve, Journal of Physiology. 1952; 117(4):500– 44. https://doi.org/10.1113/jphysiol.1952.sp004764. PMid: 12991237, PMCid: PMC1392413.

  • Izhikevich EM. Simple model of spiking neurons, IEEE Transaction Neural Networking. 2003; 14(6):1569–72.

  • Nageswaran JM, Dutt N, Krichmar JL, Nicolau A, Veidenbaum AV. A configurable simulation environment for the efficient simulation of large-scale spiking neural networks on graphics processors, Neural Networking. 2009; 22(5–6):79–800. https://doi.org/10.1016/j.neunet. 2009.06.028.

  • Fidjeland A, Roesch E, Shanahan M, Luk W. NeMo: A platform for neural modeling of spiking neurons using GPUs. 20th IEEE International Conference on Application-specific Systems, Architectures and Processors; 2009. https://doi.org/10.1109/ASAP.2009.24.

  • Garrido A, Carrillo RR, Luque NR, Ros E. Event and time driven hybrid simulation of spiking neural networks, Advances in Computational Intelligence. 2011; 6691:554– 61. https://doi.org/10.1007/978-3-642-21501-8_69.

  • Goodman D, Brette R. The Brian simulator, Front in Neuroscience. 2009; 3(2):192–97. https://doi.org/10.3389/neuro.01.026.2009. PMid: 20011141, PMCid: PMC2751620.

  • NEURON Simulator. http://www.neuron.yale.edu/neuron. Date accessed: 20/08/2018.

  • Brian Simulator. http://briansimulator.org/. Date accessed: 10/06/2018.

  • Mvaspike Simulator. http://mvaspike.gforge.inria.fr/. Date accessed: 04/04/2013.

  • NeMo Simulator. http://nemosim.sourceforge.net/. Date accessed: 10/07/2018.

  • GeNN Simulator. http:genn-team.github.io/genn/. Date accessed: 15/07/2018.


Abstract Views: 258

PDF Views: 0




  • Massively Parallel Computational Schemes for Simulating Spiking Neural Networks using GPU Accelerators

Abstract Views: 258  |  PDF Views: 0

Authors

N. Sreenivasa
Department of Computer Science and Engineering, Jain University, Nitte Meenakshi Institute of Technology, P.O. Box 6429, Yelahanka, Bengaluru – 560064, Karnataka, India
S. Balaji
Centre for Incubation, Innovation, Research and Consultancy, Jyothy Institute of Technology, Tataguni, Off Kanakapura Road, Bengaluru – 560082, Karnataka, India

Abstract


Objectives: To review various tools available for simulating Spiking Neural Networks using heterogeneous parallel processing platforms that help to reduce cost, increase the computational speed and also to document/archive lessons learnt. Methods/Statistical Analysis: The computational speed is a continuing challenge for simulating genuine spiking neural network models. Understanding of the spiking neural networks is significantly simplified by computer simulators like NEST, GeNN, EDLUT and BRIAN. Findings: Simulation is a handy toolkit of scientists and engineers of all disciplines. NEST, GeNN, EDLUT and BRIAN simulators help in achieving better performance not in terms of same kind of processing but with additional special tasks which require more computational power. BRIAN and EDLUT which are hybrid simulators supports both time driven and event driven techniques and outperform when compared to other simulators. Application/Improvements: Using BRIAN and EDLUT simulation techniques we can achieve the high performance when compared to other spiking neural simulation techniques.

References





DOI: https://doi.org/10.17485/ijst%2F2018%2Fv11i39%2F130826