An Artificial L‐ReLU Neuron with Asymmetric Diffusive Memristor for High‐Accuracy Neuromorphic Systems
Source abstract
Abstract
Abstract Artificial neurons with leaky rectified linear unit (L‐ReLU) function can effectively process negative information, enhancing the neuromorphic systems capbility to handle negative values. Memristive devices show great potential in building compact and bio‐plausible artificial neurons, however, a neuron device that supports L‐ReLU functions is still lacking. In this work, a compact L‐ReLU neuron is proposed based on a bipolar asymmetrical diffusive memristor. Utilizing intercalation and leveraging the migration diffusion ratio of Ag ions, devices meeting the characteristics of the L‐ReLU function are fabricated (Ag/TaOx/SiOx/Pt). Thus, the constructed neuron can encode the positive and negative input information into positive and negative spikes, respectively, with L‐ReLU‐like response function. In addition, to address the problem that the frequency only has positive values, a vector frequency (VF) is defined to describe the frequency of the positive and negative spikes. Moreover, a convolutional neural network and You‐Only‐Look‐Once version 3 (YOLOv3) network are constructed with the L‐ReLU neurons for gesture language recognition and target detection tasks, respectively. The network based on L‐ReLU neurons can avoid negative information squandering, showing a higher inference accuracy than the network with ReLU neurons. The results show that the neurons can offer a promising platform for building high‐accuracy neuromorphic systems.
Source: crossref · Quality: baseline
Research context
OpenAlex topics: Advanced Memory and Neural Computing · Neuroscience and Neural Engineering · Neural dynamics and brain function
SciVX discovery tags are terminology-based. OpenAlex topics are source-provided classifications.
Affiliations
School of Integrated Circuits Anhui University Hefei Anhui 230601 China
Frontier Institute of Chip and System Fudan University Shanghai 200433 China
Funding
Natural Science Foundation of Anhui Province
References (40)
Show source-provided references
1. 10.1002/adma.20200361010.1002/adma.202003610 ↗
2. 10.1002/adfm.20220574410.1002/adfm.202205744 ↗
3. 10.1038/s41467-022-30432-210.1038/s41467-022-30432-2 ↗
4. 10.1016/j.scib.2021.04.01410.1016/j.scib.2021.04.014 ↗
5. 10.1002/adfm.20230812910.1002/adfm.202308129 ↗
6. R. M.Patton C. D.Schuman S.Kulkarni M.Parsa J. P.Mitchell N. Q.Haas C.Stahl P.Date T. E.PotokS.Snyder inInternational conference on neuromorphic systems 2021 2021 pp.1–5 https://org.doi/10.1145/3477145.3477170.10.1145/3477145.3477170 ↗
7. 10.1098/rsfs.2019.012210.1098/rsfs.2019.0122 ↗
8. 10.1002/adma.20200439810.1002/adma.202004398 ↗
9. 10.1038/s41598-020-76823-710.1038/s41598-020-76823-7 ↗
10. 10.34133/icomputing.000610.34133/icomputing.0006 ↗
11. 10.1002/adma.20230106310.1002/adma.202301063 ↗
12. N.Rathi G.Srinivasan P.Panda K.Roy “Enabling deep spiking neural networks with hybrid conversion and spike timing dependent backpropagation ” arXiv preprintarXiv:2005.01807 2020.
13. 10.1038/s41586-019-1677-210.1038/s41586-019-1677-2 ↗
14. 10.1109/TETC.2024.343513510.1109/TETC.2024.3435135 ↗
15. T.Bu W.Fang J.Ding P.Dai Z.Yu T.Huang “Optimal ANN‐SNN conversion for high‐accuracy and ultra‐low‐latency spiking neural networks ” arXiv preprintarXiv:2303.04347 2023.
16. Wu Z. · IEEE Trans. Circu. Syst. II: Express Briefs · 2023 · 71 · 1446
17. 10.1109/TCDS.2023.331163410.1109/TCDS.2023.3311634 ↗
18. 10.3389/fnins.2022.90914610.3389/fnins.2022.909146 ↗
19. M.Jerry W.Tsai B.Xie X.Li V.Narayanan A.Raychowdhury S.Datta 2016 74th Annual Device Research Conference (DRC) IEEE Newark DE USA August 2016 pp.1‐2.10.1109/DRC.2016.7548503 ↗
20. X.Zhang Z.Wang W.Song R.Midya Y.Zhuo R.Wang M.Rao N. K. U.Qiangfei Xia J. J.Yang Q.Liu M.Liu 2019 IEEE International Electron Devices Meeting (IEDM) IEEE San Francisco CA USA February 2019 pp.6.7.1–6.7.4.10.1109/IEDM19573.2019.8993519 ↗
21. 10.1109/TED.2022.315246810.1109/TED.2022.3152468 ↗
22. M.Basirat P. M.Roth inProceedings of the IEEE/CVF Winter Conference on Applications of Computer Vision 2020 pp.1218–1227.
23. 10.1109/TCSII.2024.336482210.1109/TCSII.2024.3364822 ↗
24. 10.1002/adfm.20170486210.1002/adfm.201704862 ↗
25. 10.1109/TCSII.2024.338062510.1109/TCSII.2024.3380625 ↗
26. 10.1038/s41467-023-39033-z10.1038/s41467-023-39033-z ↗
27. 10.1109/LED.2020.300658110.1109/LED.2020.3006581 ↗
28. 10.1039/D2NH00536K10.1039/D2NH00536K ↗
29. 10.1038/nmat475610.1038/nmat4756 ↗
30. 10.3390/nano1317244310.3390/nano13172443 ↗
31. 10.1002/aelm.20220103810.1002/aelm.202201038 ↗
32. 10.1002/aelm.20230060810.1002/aelm.202300608 ↗
33. 10.1039/D3NH00571B10.1039/D3NH00571B ↗
34. 10.1002/aelm.20200069510.1002/aelm.202000695 ↗
35. 10.1109/JPROC.2023.330808810.1109/JPROC.2023.3308088 ↗
36. 10.1109/TED.2021.309503310.1109/TED.2021.3095033 ↗
37. 10.1038/s41467-018-07979-010.1038/s41467-018-07979-0 ↗
38. 10.1021/nn101066710.1021/nn1010667 ↗
39. 10.1007/BF0101786010.1007/BF01017860 ↗
40. D. P.Kingma J.Ba “Adam: A method for stochastic optimization ”arXiv preprint arXiv:1412.6980 2014.