CMOS + stochastic nanomagnets: heterogeneous computers for probabilistic inference and learning

Fuente: arXiv
Salvato in:
Dettagli Bibliografici
Autori principali: Singh, Nihal Sanjay, Kobayashi, Keito, Cao, Qixuan, Selcuk, Kemal, Hu, Tianrui, Niazi, Shaila, Aadit, Navid Anjum, Kanai, Shun, Ohno, Hideo, Fukami, Shunsuke, Camsari, Kerem Y.
Natura: Preprint
Pubblicazione: 2023
Soggetti:
Accesso online:
Tags: Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
_version_ 1866916180840677376
author Singh, Nihal Sanjay
Kobayashi, Keito
Cao, Qixuan
Selcuk, Kemal
Hu, Tianrui
Niazi, Shaila
Aadit, Navid Anjum
Kanai, Shun
Ohno, Hideo
Fukami, Shunsuke
Camsari, Kerem Y.
author_facet Singh, Nihal Sanjay
Kobayashi, Keito
Cao, Qixuan
Selcuk, Kemal
Hu, Tianrui
Niazi, Shaila
Aadit, Navid Anjum
Kanai, Shun
Ohno, Hideo
Fukami, Shunsuke
Camsari, Kerem Y.
contents Extending Moore's law by augmenting complementary-metal-oxide semiconductor (CMOS) transistors with emerging nanotechnologies (X) has become increasingly important. One important class of problems involve sampling-based Monte Carlo algorithms used in probabilistic machine learning, optimization, and quantum simulation. Here, we combine stochastic magnetic tunnel junction (sMTJ)-based probabilistic bits (p-bits) with Field Programmable Gate Arrays (FPGA) to create an energy-efficient CMOS + X (X = sMTJ) prototype. This setup shows how asynchronously driven CMOS circuits controlled by sMTJs can perform probabilistic inference and learning by leveraging the algorithmic update-order-invariance of Gibbs sampling. We show how the stochasticity of sMTJs can augment low-quality random number generators (RNG). Detailed transistor-level comparisons reveal that sMTJ-based p-bits can replace up to 10,000 CMOS transistors while dissipating two orders of magnitude less energy. Integrated versions of our approach can advance probabilistic computing involving deep Boltzmann machines and other energy-based learning algorithms with extremely high throughput and energy efficiency.
format Preprint
id arxiv_https___arxiv_org_abs_2304_05949
institution arXiv
publishDate 2023
record_format arxiv
spellingShingle CMOS + stochastic nanomagnets: heterogeneous computers for probabilistic inference and learning
Singh, Nihal Sanjay
Kobayashi, Keito
Cao, Qixuan
Selcuk, Kemal
Hu, Tianrui
Niazi, Shaila
Aadit, Navid Anjum
Kanai, Shun
Ohno, Hideo
Fukami, Shunsuke
Camsari, Kerem Y.
Mesoscale and Nanoscale Physics
Artificial Intelligence
Emerging Technologies
Machine Learning
Extending Moore's law by augmenting complementary-metal-oxide semiconductor (CMOS) transistors with emerging nanotechnologies (X) has become increasingly important. One important class of problems involve sampling-based Monte Carlo algorithms used in probabilistic machine learning, optimization, and quantum simulation. Here, we combine stochastic magnetic tunnel junction (sMTJ)-based probabilistic bits (p-bits) with Field Programmable Gate Arrays (FPGA) to create an energy-efficient CMOS + X (X = sMTJ) prototype. This setup shows how asynchronously driven CMOS circuits controlled by sMTJs can perform probabilistic inference and learning by leveraging the algorithmic update-order-invariance of Gibbs sampling. We show how the stochasticity of sMTJs can augment low-quality random number generators (RNG). Detailed transistor-level comparisons reveal that sMTJ-based p-bits can replace up to 10,000 CMOS transistors while dissipating two orders of magnitude less energy. Integrated versions of our approach can advance probabilistic computing involving deep Boltzmann machines and other energy-based learning algorithms with extremely high throughput and energy efficiency.
title CMOS + stochastic nanomagnets: heterogeneous computers for probabilistic inference and learning
topic Mesoscale and Nanoscale Physics
Artificial Intelligence
Emerging Technologies
Machine Learning
url https://arxiv.org/abs/2304.05949