Spike time dependent plasticity (STDP) enabled learning in spiking neural networks using domain wall based synapses and neurons
Source abstract
Abstract
We have implemented a Spiking Neural Network (SNN) architecture using a combination of spin orbit torque driven domain wall devices and transistor based peripheral circuits as both synapses and neurons. Learning in the SNN hardware is achieved both under completely unsupervised mode and partially supervised mode through mechanisms, incorporated in our spintronic synapses and neurons, that have biological plausibility, e.g., Spike Time Dependent Plasticity (STDP) and homoeostasis. High classification accuracy is obtained on the popular Iris dataset for both modes of learning.
Source: crossref · Quality: baseline
Research context
OpenAlex topics: Advanced Memory and Neural Computing · Ferroelectric and Negative Capacitance Devices · Neural Networks and Reservoir Computing
SciVX discovery tags are terminology-based. OpenAlex topics are source-provided classifications.
Affiliations
Department of Electrical Engineering, Indian Institute of Technology Delhi 1 , Delhi 110016, India
Electrical and Computer Engineering, Georgia Institute of Technology 2 , Atlanta Georgia 30332, USA
Department of Mechanical Engineering, Indian Institute of Technology Delhi 3 , Delhi 110016, India
Funding
Department of Science and Technology India · Science and Engineering Research Board
References (45)
Show source-provided references
1. J. Phys. D Appl. Phys. · 2018 · 51 · 28300110.1088/1361-6463/aac8a5 ↗
2. J. Appl. Phys. · 2018 · 124 · 11110110.1063/1.5042413 ↗
3. Nat. Commun. · 2018 · 9 · 238510.1038/s41467-018-04484-2 ↗
4. Nat. Electronics · 2019 · 2 · 29010.1038/s41928-019-0270-x ↗
5. Nature · 2017 · 547 · 42810.1038/nature23011 ↗
6. J. Appl. Phys. · 2018 · 124 · 15212210.1063/1.5042308 ↗
7. IEEE Trans. Biomed. Circuits Syst. · 2016 · 10 · 115210.1109/tbcas.2016.2525823 ↗
8. IEEE Trans. Magn. · 2018 · 54 · 110.1109/tmag.2018.2853082 ↗
9. J. Magn. Magn. Mater. · 2019 · 489 · 16543410.1016/j.jmmm.2019.165434 ↗
10. Science · 2014 · 345 · 66810.1126/science.1254642 ↗
11. IEEE Micro · 2018 · 38 · 8210.1109/mm.2018.112130359 ↗
12. ACM J. Emerg. Tech. Com. · 2019 · 15 · 110.1145/3304103 ↗
13. Front. Neurosci. · 2018 · 12 · 89110.3389/fnins.2018.00891 ↗
14. Neural Networks · 1997 · 10 · 165910.1016/s0893-6080(97)00011-7 ↗
15. Proceedings of the IEEE · 2015 · 103 · 221910.1109/jproc.2015.2496679 ↗
16. Front. Neurosci. · 2019 · 13 · 9510.3389/fnins.2019.00095 ↗
17. 2015
18. Front. Comput. Neurosci. · 2015 · 9 · 9910.3389/fncom.2015.00099 ↗
19. Phys. Rev. Appl. · 2016 · 6 · 06400310.1103/physrevapplied.6.064003 ↗
20. Sci. Adv. · 2019 · 5 · eaau817010.1126/sciadv.aau8170 ↗
21. Appl. Phys. Rev. · 2017 · 4 · 04110510.1063/1.5012763 ↗
22. J. Neurosci. · 1998 · 18 · 1046410.1523/jneurosci.18-24-10464.1998 ↗
23. 2005
24. Nat. Rev. Neurosci. · 2003 · 4 · 88510.1038/nrn1248 ↗
25. J. Appl. Phys. · 2018 · 124 · 15212710.1063/1.5042452 ↗
26. IEEE Access · 2019 · 7 · 503410.1109/access.2018.2886854 ↗
27. J. Phys. D Appl. Phys. · 2019 · 52 · 49LT0110.1088/1361-6463/ab4157 ↗
28. AIP Adv. · 2014 · 4 · 10713310.1063/1.4899186 ↗
29. Ann. Eugenics · 1936 · 7 · 17910.1111/j.1469-1809.1936.tb02137.x ↗
30. A. Biswas, S. Prasad, S. Lashkare, and U. Ganguly, arXiv:1612.02233 (2016).
31. 2017
32. Proc. SPIE · 2019 · 11090 · 110903I10.1117/12.2530308 ↗
33. Nat. Mater. · 2013 · 12 · 61110.1038/nmat3675 ↗
34. Nat. Nanotechnol. · 2013 · 8 · 52710.1038/nnano.2013.102 ↗
35. Sci. Rep. · 2015 · 5 · 1182310.1038/srep11823 ↗
36. Nat. Mater. · 2011 · 10 · 41910.1038/nmat3020 ↗
37. Phys. Rev. B · 2014 · 90 · 18442710.1103/physrevb.90.184427 ↗
38. J. Appl. Phys. · 2014 · 115 · 21390910.1063/1.4881778 ↗
39. Phys. Rev. B · 2015 · 91 · 01443310.1103/physrevb.91.014433 ↗
40. Materials Today · 2006 · 9 · 3610.1016/s1369-7021(06)71693-5 ↗
41. Nat. Mater. · 2010 · 9 · 72110.1038/nmat2804 ↗
42. Neural Comput. · 2007 · 19 · 258110.1162/neco.2007.19.10.2581 ↗
43. 2009
44. 1988
45. IEEE Trans. Nanotechnol. · 2013 · 12 · 28810.1109/tnano.2013.2250995 ↗