Abstract
We discuss a convolutional neural network for handwritten digit classification and its hardware acceleration as an inference engine using nanoscale memristive devices in the spike domain. We study the impact of device programming variability on the spiking neural network’s (SNN) inference accuracy and benchmark its performance with an equivalent artificial neural network (ANN). We demonstrate optimization strategies to implement these networks with memristive devices with an on-off ratio as low as 10 and only 32 levels of resolution. Further, close to baseline accuracies can be maintained for the networks even if such memristive devices are used to duplicate the pre-determined kernel weights to enable parallel execution of the convolution operation.
Keywords
This is a preview of subscription content, log in via an institution.
Buying options
Tax calculation will be finalised at checkout
Purchases are for personal use only
Learn about institutional subscriptionsReferences
Hubel, D., Wiesel, T.: Receptive fields and functional architecture of monkey striate cortex. J. Physiol. 195(1), 215–243 (1968)
Lecun, Y., et al.: Gradient-based learning applied to document recognition. Proc. IEEE 86(11), 2278–2324 (1998)
Krizhevsky, A., Sutskever, I., Hinton, G.: Imagenet classification with deep convolutional neural networks. In: Advances in neural information processing systems, pp. 1097–1105 (2012)
Szegedy, C., et al.: Inception-v4, Inception-ResNet and the impact of residual connections on learning. In: AAAI, vol. 4, p. 12 (2017)
Maass, W.: Networks of spiking neurons: the third generation of neural network models. Neural Netw. 10(9), 1659–1671 (1997)
Wang, B., et al.: Firing frequency maxima of fast-spiking neurons in human, monkey, and mouse neocortex. Front. Cell. Neurosci. 10, 239 (2016). 27803650 [pmid]
Han, B., Sengupta, A., Roy, K.: On the energy benefits of spiking deep neural networks: a case study. In: 2016 International Joint Conference on Neural Networks (IJCNN), pp. 971–976. IEEE (2016)
Merolla, P.A., et al.: A million spiking-neuron integrated circuit with a scalable communication network and interface. Science 345(6197), 668–673 (2014)
Qiao, N., et al.: A reconfigurable on-line learning spiking neuromorphic processor comprising 256 neurons and 128k synapses. Front. Neurosci. 9, 141 (2015)
Davies, M., et al.: Loihi: a neuromorphic manycore processor with on-chip learning. IEEE Micro 38(1), 82–99 (2018)
Kim, S., et al.: NVM neuromorphic core with 64k-cell (256-by-256) phase change memory synaptic array with on-chip neuron circuits for continuous In-Situ learning. In: 2015 IEEE International Electron Devices Meeting (IEDM), pp. 17.1.1–17.1.4, December 2015
Burr, G.W., et al.: Neuromorphic computing using non-volatile memory. Adv. Phys. X 2(1), 89–124 (2017)
Rajendran, B., Alibart, F.: Neuromorphic computing based on emerging memory technologies. IEEE J. Emerg. Sel. Top. Circ. Syst. 6(2), 198–211 (2016)
Kuzum, D., Yu, S., Wong, P.: Synaptic electronics: materials, devices and applications. Nanotechnology 24(38), 382001 (2013)
Jo, S.H., et al.: Nanoscale memristor device as synapse in neuromorphic systems. Nano Lett. 10(4), 1297–1301 (2010). PMID: 20192230
Jackson, B.L., et al.: Nanoscale electronic synapses using phase change devices. J. Emerg. Technol. Comput. Syst. 9(2), 12 (2013)
Burr, G.W., et al.: Large-scale neural networks implemented with non-volatile memory as the synaptic weight element: comparative performance analysis (accuracy, speed, and power). In: 2015 IEEE International Electron Devices Meeting (IEDM), pp. 4.4.1–4.4.4, December 2015
Song, L., et al.: PipeLayer: a pipelined ReRAM-based accelerator for deep learning. In: 2017 IEEE International Symposium on High Performance Computer Architecture (HPCA), pp. 541–552, February 2017
Yakopcic, C., Alom, Z., Taha, T.: Memristor crossbar deep network implementation based on a convolutional neural network. In: International Joint Conference on Neural Networks (2016)
Yakopcic, C., Alom, Z., Taha, T.: Extremely parallel memristor crossbar architecture for convolutional neural network implementation. In: 2017 International Joint Conference on Neural Networks (IJCNN), pp. 1696–1703. IEEE (2017)
Chen, P.Y., et al.: Mitigating effects of non-ideal synaptic device characteristics for on-chip learning. In: 2015 IEEE/ACM International Conference on Computer-Aided Design (ICCAD), November 2015
Babu, A.V., Rajendran, B.: Stochastic deep learning in memristive networks. In: 2017 24th IEEE International Conference on Electronics, Circuits and Systems (ICECS), pp. 214–217, December 2017
Abbott, L.: Lapicque’s introduction of the integrate-and-fire model neuron (1907). Brain Res. Bull. 50, 303–304 (1999)
Kulkarni, S.R., Alexiades, J.M., Rajendran, B.: Learning and real-time classification of hand-written digits with spiking neural networks. In: 2017 24th IEEE International Conference on Electronics, Circuits and Systems (ICECS), pp. 128–131, December 2017
Calderón, A., Roa, S., Victorino, J.: Handwritten digit recognition using convolutional neural networks and Gabor filters. In: Proceedings of International Congress on Computational Intelligence (2003)
Anwani, N., Rajendran, B.: NormAD - normalized approximate descent based supervised learning rule for spiking neurons. In: International Joint Conference on Neural Networks, pp. 1–8, July 2015
Schreiber, S., et al.: A new correlation-based measure of spike timing reliability. Neurocomputing 52, 925–931 (2003)
Kulkarni, S.R., Rajendran, B.: Spiking neural networks for handwritten digit recognition-supervised learning and network optimization. Neural Netw. 103, 118–127 (2018)
Stromatias, E., et al.: Robustness of spiking deep belief networks to noise and reduced bit precision of neuro-inspired hardware platforms. Front. Neurosci. 9, 222 (2015)
Suri, M., et al.: Phase change memory as synapse for ultra-dense neuromorphic systems: application to complex visual pattern extraction. In: 2011 International Electron Devices Meeting, pp. 4.4.1–4.4.4, December 2011
Gokmen, T., Onen, M., Haensch, W.: Training deep convolutional neural networks with resistive cross-point devices. arXiv preprint arXiv:1705.08014 (2017)
Garbin, D., et al.: HfO2-based OxRAM devices as synapses for convolutional neural networks. IEEE Trans. Electron Devices 62(8), 2494–2501 (2015)
Lim, S., et al.: Adaptive learning rule for hardware-based deep neural networks using electronic synapse devices. ArXiv e-prints arXiv:1707.06381v2, July 2017
Boybat, I., et al.: Neuromorphic computing with multi-memristive synapses. ArXiv e-prints, November 2017
Panwar, N., Rajendran, B., Ganguly, U.: Arbitrary spike time dependent plasticity (STDP) in memristor by analog waveform engineering. IEEE Electron Device Lett. 38(6), 740–743 (2017)
Acknowledgments
This research was supported in part by the CAMPUSENSE project grant from CISCO Systems Inc, the Semiconductor Research Corporation (2016-SD-2717), and the National Science Foundation grant 1710009.
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2018 Springer Nature Switzerland AG
About this paper
Cite this paper
Kulkarni, S.R., Babu, A.V., Rajendran, B. (2018). Acceleration of Convolutional Networks Using Nanoscale Memristive Devices. In: Pimenidis, E., Jayne, C. (eds) Engineering Applications of Neural Networks. EANN 2018. Communications in Computer and Information Science, vol 893. Springer, Cham. https://doi.org/10.1007/978-3-319-98204-5_20
Download citation
DOI: https://doi.org/10.1007/978-3-319-98204-5_20
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-319-98203-8
Online ISBN: 978-3-319-98204-5
eBook Packages: Computer ScienceComputer Science (R0)