Training of a Spiking Neural Network with a Consideration of Memristive Crossbar Array Characteristics
- Autores: Dudkin A.P.1, Ryndin E.A.1, Andreeva N.V.1
-
Afiliações:
- Ulyanov (Lenin) St. Petersburg State Electrotechnical University “LETI”, St. Petersburg, Russia
- Edição: Volume 54, Nº 4 (2025)
- Páginas: 310-322
- Seção: NEUROMORPHIC SYSTEMS
- URL: https://rjsvd.com/0544-1269/article/view/690996
- DOI: https://doi.org/10.31857/S0544126925040058
- EDN: https://elibrary.ru/qhfrkq
- ID: 690996
Citar
Texto integral



Resumo
A model, methodology and software tools for modeling a spiking neural network in the training mode taking into account the peculiarities of the functioning of memristive crossbar arrays have been developed. The influence of voltage drops on interconnections, discrete step of tuning of conductivity levels of memristive elements and nonlinearity of their volt-ampere characteristics on the efficiency of execution of spiking neural network training algorithms has been investigated. The results of testing the spiking neural network in the training mode and inference mode in the task of image recognition with the use of the developed modeling technique taking into account the characteristics of experimentally manufactured memristive structures have been obtained.
Palavras-chave
Sobre autores
A. Dudkin
Ulyanov (Lenin) St. Petersburg State Electrotechnical University “LETI”, St. Petersburg, Russia
Autor responsável pela correspondência
Email: anddudkin000@gmail.com
St. Petersburg, Russia
E. Ryndin
Ulyanov (Lenin) St. Petersburg State Electrotechnical University “LETI”, St. Petersburg, Russia
Email: rynenator@gmail.com
St. Petersburg, Russia
N. Andreeva
Ulyanov (Lenin) St. Petersburg State Electrotechnical University “LETI”, St. Petersburg, Russia
Email: nvandr@gmail.com
St. Petersburg, Russia
Bibliografia
- Mutlu O., Ghose S., Gómez-Luna J., Ausavarungnirun R. Processing data where it makes sense: Enabling in-memory computation // Microprocessors and Microsystems. 2019. V. 67. P. 28–41.
- Rojas R. The Backpropagation Algorithm // Neural Networks. Springer, Berlin, Heidelberg. 1996. P. 149–182.
- Caporale N., Dan Y. Spike timing–dependent plasticity: a Hebbian learning rule // Annual Review of Neuroscience. 2008. V. 31. N. 1. P. 25–46.
- Li X., Tang J., Zhang Q., Gao B., Yang J.J., Song S., Wu W., Zhang W., Yao P., Deng N. et al. Power-efficient neural network with artificial dendrites // Nature Nanotechnology. 2020. V. 15. N. 9. P. 776–782.
- Li Y., Su K., Chen H., Zou X., Wang C., Man H., Liu K., Xi X., Li T. Research progress of neural synapses based on memristors // Electronics. 2023. V. 12. N. 15. 3298.
- Surazhevsky I.A., Demin V.A., Ilyasov A.I., Emelyanov A.V., Nikiruy K.E., Rylkov V.V., Shchanikov S.A., Bordanov I.A., Gerasimova S.A., Guseinov D.V. et al. Noise-assisted persistence and recovery of memory state in a memristive spiking neuromorphic network // Chaos, Solitons & Fractals. 2021. V. 146. 110890.
- Huang J., Serb A., Stathopoulos S., Prodromakis T. Text classification in memristor-based spiking neural networks // Neuromorphic Computing and Engineering. 2023. V. 3. N. 1. 014003.
- Guo Y., Wu H., Gao B., Qian H. Unsupervised learning on resistive memory array based spiking neural networks // Frontiers in neuroscience. 2019. V. 13. 812.
- Prezioso M., Mahmoodi M.R., Bayat F.M., Nili H., Kim H., Vincent A., Strukov D.B. Spike-timing-dependent plasticity learning of coincidence detection with passively integrated memristive circuits // Nature communications. 2018. V. 9. N. 1. 5311.
- Milo V., Pedretti G., Carboni R., Calderoni A., Ramaswamy N., Ambrogio S., Ielmini D. Demonstration of hybrid CMOS/RRAM neural networks with spike time/rate-dependent plasticity // 2016 IEEE International Electron Devices Meeting (IEDM). 2016. P. 16.8.1–16.8.4.
- Kim H., Mahmoodi M., Nili H., Strukov D.B. 4K-memristor analog-grade passive crossbar circuit // Nature communications. 2021. V. 12. N. 1. 5198.
- Andreeva N., Ivanov A., Petrov A. Multilevel resistive switching in TiO2/Al2O3 bilayers at low temperature // AIP Advances. 2018. V. 8. N. 2. 025208.
- Rao M., Tang H., Wu J., Song W., Zhang M., Yin W., Zhuo Y., Kiani F., Chen B., Jiang X. et al. Thousands of conductance levels in memristors integrated on CMOS // Nature. 2023. V. 615. N. 7954. P. 823–829.
- Cai F., Correll J.M., Lee S.H., Lim Y., Bothra V., Zhang Z., Flynn M.P., Lu W.D. A fully integrated reprogrammable memristor–CMOS system for efficient multiply–accumulate operations // Nature electronics. 2019. V. 2. N. 7. P. 290–299.
- Yao P., Wu H., Gao B., Tang J., Zhang Q., Zhang W., Yang J.J., Qian H. Fully hardware-implemented memristor convolutional neural network // Nature. 2020. V. 577. N. 7792. P. 641–646.
- Kiani F., Yin J., Wang Z., Yang J.J., Xia Q. A fully hardware-based memristive multilayer neural network // Science advances. 2021. V. 7. N. 48. eabj4801.
- Yin S., Sun X., Yu S., Seo J.S. High-throughput in-memory computing for binary deep neural networks with monolithically integrated RRAM and 90-nm CMOS // IEEE Transactions on Electron Devices. 2020. V. 67. N. 10. P. 4185–4192.
- Li C., Ignowski J., Sheng X., Wessel R., Jaffe B., Ingemi J., Graves C., Strachan J.P. CMOS-integrated nanoscale memristive crossbars for CNN and optimization acceleration // 2020 IEEE International Memory Workshop (IMW). 2020. P. 1–4.
- Pedretti G., Mannocci P., Li C., Sun Z., Strachan J.P., Ielmini D. Redundancy and analog slicing for precise in-memory machine learning—Part I: Programming techniques // IEEE Transactions on Electron Devices. 2021. V. 68. N. 9. P. 4373–4378.
- Chen W.H., Dou C., Li K.X., Lin W.Y., Li P.Y., Huang J.H., Wang J.H., Wei W.C., Xue C.X., Chiu Y.C. et al. CMOS-integrated memristive non-volatile computing-in-memory for AI edge processors // Nature Electronics. 2019. V. 2. N. 9. P. 420–428.
- Hirtzlin T., Bocquet M., Penkovsky B., Klein J.O., Nowak E., Vianello E., Portal J.M., Querlioz D. Digital biologically plausible implementation of binarized neural networks with differential hafnium oxide resistive memory arrays // Frontiers in neuroscience. 2020. V. 13. 1383.
- Mochida R., Kouno K., Hayata Y., Nakayama M., Ono T., Suwa H., Yasuhara R., Katayama K., Mikawa T., Gohou Y. A 4M synapses integrated analog ReRAM based 66.5 TOPS/W neural-network processor with cell current controlled writing and flexible network architecture // 2018 IEEE Symposium on VLSI Technology. 2018. P. 175–176.
- Su F., Chen W.H., Xia L., Lo C.P., Tang T., Wang Z., Hsu K.H., Cheng M., Li J.Y., Xie Y. et al. A 462GOPs/J RRAM-based nonvolatile intelligent processor for energy harvesting IoE system featuring nonvolatile logics and processing-in-memory // 2017 Symposium on VLSI Technology. 2017. P. T260–T261.
- Ambrogio S., Narayanan P., Tsai H., Shelby R.M., Boybat I., Di Nolfo C., Sidler S., Giordano M., Bodini M., Farinha N.C.P. et al. Equivalent-accuracy accelerated neural-network training using analogue memory // Nature. 2018. V. 558. N. 7708. P. 60–67.
- Deng L. The MNIST database of handwritten digit images for machine learning research [best of the web] // IEEE signal processing magazine. 2012. V. 29. N. 6. P. 141–142.
- Dudkin A.P., Ryndin E.A., Andreeva N.V. Modeling the Functional Features of a Memristive Crossbar Array in Neuromorphic Electronic Modules // Russian Microelectronics. 2024. V. 53. N. 6. P. 554–566.
Arquivos suplementares
