141st CEMS Colloquium

Title

Probabilistic Computing with Stochastic Spintronic Devices

Abstract

Progress of computing has mainly relied on transistor scaling for a long time following the Moore’s law, but rising cost per device has disrupted this economical model. As a result, the field is shifting toward domain-specific accelerators and the use of novel devices suitable for specific tasks. Spintronic devices are expected to play a pivotal role in this shift. Beyond their role in nonvolatile memory, their ability to generate random numbers with a small power and a large output signal enables scalable probabilistic computers that tackle computationally hard problems. In this talk, basic principles of the spintronic probabilistic computing and some proof-of-concepts for combinatorial optimization [1], machine learning [2], and quantum simulation [3] will be shown. The device physics and engineering [4-6] enhancing the performance of the computer will be also discussed.

[1] W. A. Borders et al., Nature 573, 390 (2019).
[2] J. Kaiser et al., Phys. Rev. Appl. 17, 014016 (2022).
[3] A. Grimardi et al., IEEE IEDM 2022, 22.4 (2022).
[4] K. Hayakawa et al., Phys. Rev. Lett. 126, 117202 (2021).
[5] S. Kanai et al., Phys. Rev. B 103, 094423 (2021).
[6] K. Kobayashi et al., Phys. Rev. Appl., 18, 054085 (2022).

Contact

email: citonaga@riken.jp
If you are not affiliated with RIKEN and plan to attend the Colloquium, please provide full name and contact addresses as pre-registration is required.