Demonstration of in-plane magnetized stochastic magnetic tunnel junction for binary stochastic neuron
Source abstract
Abstract
A binary stochastic neuron (BSN) or a probabilistic bit (p-bit) randomly fluctuates between digitized “0” and “1” with a controllable functionality of time-averaged value. Such an unconventional bit is the most essential building block for the recently proposed stochastic neural networks and probabilistic computing. Here, we experimentally implement a magnetic tunnel junction (MTJ) for BSN, with relaxation times on the order of tens of milliseconds that can be modulated by a current-induced spin-transfer torque. The NIST Statistical Test Suite (800-22a) is used to verify true random number generation by the BSN-MTJ device. Our results suggest the possibility of using the artificial BSN MTJ device in neuromorphic applications as well as in a recently proposed probabilistic computing.
Source: crossref · Quality: baseline
Research context
OpenAlex topics: Advanced Memory and Neural Computing · Ferroelectric and Negative Capacitance Devices · Neural Networks and Applications
SciVX discovery tags are terminology-based. OpenAlex topics are source-provided classifications.
Affiliations
Center for Spintronics, Korea Institute of Science and Technology 1 , Seoul 02792, South Korea
University of Science and Technology of Korea (UST) 2 , Daejeon 34113, South Korea
KU-KIST Graduate School, Korea University 3 , Seoul, South Korea
Funding
Korea Institute of Science and Technology · National Research Foundation of Korea
References (25)
Show source-provided references
1. Nature · 2017 · 550 · 35410.1038/nature24270 ↗
2. IEEE Micro · 2018 · 38 · 8210.1109/mm.2018.112130359 ↗
3. Nature · 2019 · 575 · 60710.1038/s41586-019-1677-2 ↗
4. Int. J. Theor. Phys. · 1982 · 21 · 46710.1007/bf02650179 ↗
5. Nature · 2019 · 574 · 50510.1038/s41586-019-1666-5 ↗
6. Science · 2018 · 360 · 19510.1126/science.aao4309 ↗
7. Science · 2018 · 362 · 30810.1126/science.aar3106 ↗
8. Nature · 2019 · 573 · 39010.1038/s41586-019-1557-9 ↗
9. Appl. Phys. Rev. · 2019 · 6 · 01130510.1063/1.5055860 ↗
10. Proc. IEEE · 1990 · 78 · 162910.1109/5.58356 ↗
11. Phys. Rev. Lett. · 1995 · 74 · 327310.1103/physrevlett.74.3273 ↗
12. Mater. Today · 2017 · 20 · 53010.1016/j.mattod.2017.07.007 ↗
13. Phys. Rev. Appl. · 2019 · 12 · 05405610.1103/physrevapplied.12.054056 ↗
14. Phys. Rev. B · 2021 · 103 · 09442310.1103/physrevb.103.094423 ↗
15. Nano Lett. · 2021 · 21 · 204010.1021/acs.nanolett.0c04652 ↗
16. Phys. Rev. Lett. · 2021 · 126 · 11720210.1103/physrevlett.126.117202 ↗
17. Phys. Rev. B · 2004 · 69 · 21440910.1103/physrevb.69.214409 ↗
18. IEEE Trans. Electron Devices · 2019 · 66 · 535310.1109/ted.2019.2948218 ↗
19. Appl. Phys. Lett. · 2017 · 111 · 24310710.1063/1.5012091 ↗
20. 2010
21. Phys. Rev. Appl. · 2017 · 8 · 05404510.1103/physrevapplied.8.054045 ↗
22. IEEE Trans. Circuits Syst. · 2022 · 69 · 27210.1109/tcsi.2021.3087512 ↗
23. Phys. Rev. B · 2011 · 84 · 06443910.1103/physrevb.84.064439 ↗
24. J. Magn. Magn. Mater. · 2021 · 521 · 16750610.1016/j.jmmm.2020.167506 ↗
25. Phys. Rev. Appl. · 2021 · 15 · 06404610.1103/physrevapplied.15.064046 ↗