Skip to main navigation Skip to search Skip to main content

Spin-Orbit Torque-Driven Markov Chain Carrier

  • Min Wang
  • , Jinhao Li
  • , Hongchao Zhang
  • , Yuanfu Zhao
  • , Zhaohao Wang
  • Beihang University
  • Truth Memory Coporation
  • Beijing Microelectronics Technology Institute

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

Abstract

Markov chain (MC), as the foundational statistical tool, enables important applications ranging from stochastic sampling to advanced computation. Spin-orbit torque (SOT) devices feature probabilistic behaviors, which are suitable as a physical carrier for implementing MC designs. Our work demonstrates the SOT-driven state transitions and the memoryless nature of the proposed design. The high-dimensional states transition based on the SOT-MC design is analyzed. Our work paves a new path for high-performance computing systems.

Original languageEnglish
Title of host publicationInternational SoC Design Conference 2025, ISOCC 2025 - Proceedings of Technical Papers
PublisherInstitute of Electrical and Electronics Engineers Inc.
ISBN (Electronic)9798331586423
DOIs
StatePublished - 2025
Event22nd International SoC Design Conference, ISOCC 2025 - Busan, Korea, Republic of
Duration: 15 Oct 202518 Oct 2025

Publication series

NameInternational SoC Design Conference 2025, ISOCC 2025 - Proceedings of Technical Papers

Conference

Conference22nd International SoC Design Conference, ISOCC 2025
Country/TerritoryKorea, Republic of
CityBusan
Period15/10/2518/10/25

Keywords

  • Markov Chain
  • Probabilistic Behavior
  • SOT

Fingerprint

Dive into the research topics of 'Spin-Orbit Torque-Driven Markov Chain Carrier'. Together they form a unique fingerprint.

Cite this