All-Spin Bayesian Neural Networks

Kezhou Yang, Akul Malhotra, Sen Lu, Abhronil Sengupta

Research output: Contribution to journalArticle

1 Scopus citations

Abstract

Probabilistic machine learning enabled by the Bayesian formulation has recently gained significant attention in the domain of automated reasoning and decision-making. While impressive strides have been recently made to scale up the performance of deep Bayesian neural networks, they have been primarily standalone software efforts without any regard to the underlying hardware implementation. In this article, we propose an 'all-spin' Bayesian neural network where the underlying spintronic hardware provides a better match to the Bayesian computing models. To the best of our knowledge, this is the first exploration of a Bayesian neural hardware accelerator enabled by emerging post-CMOS technologies. We develop an experimentally calibrated device-circuit-algorithm cosimulation framework and demonstrate 24× reduction in energy consumption against an iso-network CMOS baseline implementation.

Original languageEnglish (US)
Article number8994189
Pages (from-to)1340-1347
Number of pages8
JournalIEEE Transactions on Electron Devices
Volume67
Issue number3
DOIs
StatePublished - Mar 2020

All Science Journal Classification (ASJC) codes

  • Electronic, Optical and Magnetic Materials
  • Electrical and Electronic Engineering

Fingerprint Dive into the research topics of 'All-Spin Bayesian Neural Networks'. Together they form a unique fingerprint.

  • Cite this