Abstract
A composite memristor-diode crossbar, which is the basis of the memory matrix of a biomorphic neuroprocessor, is fabricated using the methods of magnetron sputtering and electron lithography. Hardware testing of the memory-matrix operation in the mode of single-layer perceptron synapses, which can be considered as the first layer of a biomorphic neural network and perform the primary processing of incoming information in a biomorphic neuroprocessor, is carried out. The generation of a new association during retraining caused by the arrival of new input information is demonstrated. The influence of adjacent neurons on the synaptic current due to parasitic currents between adjacent cells in the memristor-diode crossbar is shown. With the help of the constructed hardware spiking neural network, the arrival of new unknown information can be identified with the generation of new associations in a biomorphic neural processor, and the neural network can learn to comprehend this information while improving itself. Therefore, it will allow the transition from narrow to general artificial intelligence.
Similar content being viewed by others
REFERENCES
L. Khacef, N. Abderrahmane, and B. Miramond, in Proceedings of the International Joint Conference on Neural Network IJCNN'2018, 2018, p. 1. https://doi.org/10.1109/ijcnn.2018.8489241
J. L. Lobo, J. D. Ser, A. Bifet, and N. Kasabov, Neural Netw. 121, 88 (2020). https://doi.org/10.1016/j.neunet.2019.09.004
M. Prezioso, F. Merrikh-Bayat, D. B. Strukov, et al., Nature (London, U.K.) 521, 61 (2015). https://doi.org/10.1038/nature14441
Y. V. Pershin and M. di Ventra, Neural Netw. 23, 881 (2010). https://doi.org/10.1016/j.neunet.2010.05.001
Z. Wang, X. Wang, and A. Novel, IEEE Trans. Circuits Syst. I: Reg. Pap. 65, 2210 (2018). https://doi.org/10.1109/TCSI.2017.2780826
L. Yang, Z. Zeng, Y. Huang, and S. Wen, IEEE Trans. Cogn. Dev. Syst. 10, 1133 (2018). https://doi.org/10.1109/TCDS.2018.2859303
Z. Wang, M. Rao, J.-W. Han, et al., Nat. Commun. 9, 3208 (2018). https://doi.org/10.1038/s41467-018-05677-5
X. Zhang and K. Long, IEEE Access. 7, 15262 (2019). https://doi.org/10.1109/ACCESS.2019.2894634
A. A. Minnekhanov, A. V. Emelyanov, D. A. Lapkin, et al., Sci. Rep. 9, 10800 (2019). https://doi.org/10.1038/s41598-019-47263-9
V. A. Demin, D. V. Nekhaev, I. A. Surazhevsky, et al., Neural Netw. 134, 64 (2021). https://doi.org/10.1016/j.neunet.2020.11.005
M. Hu, C. E. Graves, C. Li, et al., Adv. Mater. 30, 1705914 (2018). https://doi.org/10.1002/adma.201705914
F. Cai, S. Kumar, and T. V. Vaerenbergh, Nat. Electron. 3, 409 (2020). https://doi.org/10.1038/s41928-020-0436-6
A. D. Pisarev, A. N. Busygin, S. Yu. Udovichenko, and O. V. Maevsky, Microelectron. J. 102, 104827 (2020). https://doi.org/10.1016/j.mejo.2020.104827
V. A. Filippov, A. N. Bobylev, A. N. Busygin, et al., Neural Comput. Appl. 32, 2471 (2020). https://doi.org/10.1007/s00521-019-04383-7
J. Liu, L. J. Mcdaid, J. Harkin, et al., IEEE Trans. Neural Netw. Learn. Syst. 30, 865 (2018). https://doi.org/10.1109/TNNLS.2018.2854291
A. Pisarev, A. Busygin, A. Bobylev, et al., Microelectron. Eng. 236, 111471 (2021). https://doi.org/10.1016/j.mee.2020.111471
A. N. Bobylev and S. Yu. Udovichenko, Russ. Microelectron. 45, 396 (2016). https://doi.org/10.1134/S1063739716060020
A. D. Pisarev, A. N. Busygin, S. Yu. Udovichenko, and O. V. Maevsky, Microelectron. Eng. 198, 1 (2018). https://doi.org/10.1016/j.mee.2018.06.008
Funding
This work was supported by the Russian Foundation for Basic Research (grants no. 19-07-00272 “Electrophysical properties of a combined memristor-diode crossbar—a new component of nanoelectronics intended for the manufacture of memory and logical matrices of a neuroprocessor” and no. 19-37-90030 “Generation of new knowledge in a neural network based on an array of memristor synapses in the memory matrix of a biomorphic neuroprocessor and the principles of increasing the speed and energy efficiency of information processing on a specialized device in comparison with existing computing facilities”).
Author information
Authors and Affiliations
Corresponding author
Rights and permissions
About this article
Cite this article
Bobylev, A.N., Busygin, A.N., Gubin, A.A. et al. Manufacture and Testing of a Pulsed Hardware Neural Network with Memristor Synapses for a Biomorphic Neuroprocessor. Nanotechnol Russia 16, 761–766 (2021). https://doi.org/10.1134/S2635167621060057
Received:
Revised:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1134/S2635167621060057