Automatic Extraction and Compensation of P-Bit Device Variations in Large Array Utilizing Boltzmann Machine Training

A Probabilistic Bit (P-Bit) device serves as the core hardware for implementing Ising computation. However, the severe intrinsic variations of stochastic P-Bit devices hinder the large-scale expansion of the P-Bit array, significantly limiting the practical usage of Ising computation. In this work,...

Full description

Saved in:
Bibliographic Details
Main Authors: Bolin Zhang, Yu Liu, Tianqi Gao, Jialiang Yin, Zhenyu Guan, Deming Zhang, Lang Zeng
Format: Article
Language:English
Published: MDPI AG 2025-01-01
Series:Micromachines
Subjects:
Online Access:https://www.mdpi.com/2072-666X/16/2/133
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1849719135374671872
author Bolin Zhang
Yu Liu
Tianqi Gao
Jialiang Yin
Zhenyu Guan
Deming Zhang
Lang Zeng
author_facet Bolin Zhang
Yu Liu
Tianqi Gao
Jialiang Yin
Zhenyu Guan
Deming Zhang
Lang Zeng
author_sort Bolin Zhang
collection DOAJ
description A Probabilistic Bit (P-Bit) device serves as the core hardware for implementing Ising computation. However, the severe intrinsic variations of stochastic P-Bit devices hinder the large-scale expansion of the P-Bit array, significantly limiting the practical usage of Ising computation. In this work, a behavioral model which attributes P-Bit variations to two parameters, <inline-formula><math xmlns="http://www.w3.org/1998/Math/MathML" display="inline"><semantics><mi>α</mi></semantics></math></inline-formula> and <inline-formula><math xmlns="http://www.w3.org/1998/Math/MathML" display="inline"><semantics><mrow><mo>Δ</mo><mi>V</mi></mrow></semantics></math></inline-formula>, is proposed. Then the weight compensation method is introduced, which can mitigate <inline-formula><math xmlns="http://www.w3.org/1998/Math/MathML" display="inline"><semantics><mi>α</mi></semantics></math></inline-formula> and <inline-formula><math xmlns="http://www.w3.org/1998/Math/MathML" display="inline"><semantics><mrow><mo>Δ</mo><mi>V</mi></mrow></semantics></math></inline-formula> of P-Bit device variations by rederiving the weight matrix, enabling them to compute as ideal identical P-Bits without the need for weights retraining. Accurately extracting the <inline-formula><math xmlns="http://www.w3.org/1998/Math/MathML" display="inline"><semantics><mi>α</mi></semantics></math></inline-formula> and <inline-formula><math xmlns="http://www.w3.org/1998/Math/MathML" display="inline"><semantics><mrow><mo>Δ</mo><mi>V</mi></mrow></semantics></math></inline-formula> simultaneously from a large P-Bit array which is prerequisite for the weight compensation method is a crucial and challenging task. To solve this obstacle, we present the novel automatic variation extraction algorithm which can extract device variations of each P-Bit in a large array based on Boltzmann machine learning. In order for the accurate extraction of variations from an extendable P-Bit array, an Ising Hamiltonian based on a 3D ferromagnetic model is constructed, achieving precise and scalable array variation extraction. The proposed Automatic Extraction and Compensation algorithm is utilized to solve both 16-city traveling salesman problem (TSP) and 21-bit integer factorization on a large P-Bit array with variation, demonstrating its accuracy, transferability, and scalability.
format Article
id doaj-art-4d0d0f45d0b24fd1b466acdefb5eedb9
institution DOAJ
issn 2072-666X
language English
publishDate 2025-01-01
publisher MDPI AG
record_format Article
series Micromachines
spelling doaj-art-4d0d0f45d0b24fd1b466acdefb5eedb92025-08-20T03:12:12ZengMDPI AGMicromachines2072-666X2025-01-0116213310.3390/mi16020133Automatic Extraction and Compensation of P-Bit Device Variations in Large Array Utilizing Boltzmann Machine TrainingBolin Zhang0Yu Liu1Tianqi Gao2Jialiang Yin3Zhenyu Guan4Deming Zhang5Lang Zeng6National Key Laboratory of Spintronics, Hangzhou International Innovation Institute, Beihang University, Hangzhou 311115, ChinaNational Key Laboratory of Spintronics, Hangzhou International Innovation Institute, Beihang University, Hangzhou 311115, ChinaNational Key Laboratory of Spintronics, Hangzhou International Innovation Institute, Beihang University, Hangzhou 311115, ChinaNational Key Laboratory of Spintronics, Hangzhou International Innovation Institute, Beihang University, Hangzhou 311115, ChinaSchool of Cyber Science and Technology, Beihang University, Beijing 100191, ChinaNational Key Laboratory of Spintronics, Hangzhou International Innovation Institute, Beihang University, Hangzhou 311115, ChinaNational Key Laboratory of Spintronics, Hangzhou International Innovation Institute, Beihang University, Hangzhou 311115, ChinaA Probabilistic Bit (P-Bit) device serves as the core hardware for implementing Ising computation. However, the severe intrinsic variations of stochastic P-Bit devices hinder the large-scale expansion of the P-Bit array, significantly limiting the practical usage of Ising computation. In this work, a behavioral model which attributes P-Bit variations to two parameters, <inline-formula><math xmlns="http://www.w3.org/1998/Math/MathML" display="inline"><semantics><mi>α</mi></semantics></math></inline-formula> and <inline-formula><math xmlns="http://www.w3.org/1998/Math/MathML" display="inline"><semantics><mrow><mo>Δ</mo><mi>V</mi></mrow></semantics></math></inline-formula>, is proposed. Then the weight compensation method is introduced, which can mitigate <inline-formula><math xmlns="http://www.w3.org/1998/Math/MathML" display="inline"><semantics><mi>α</mi></semantics></math></inline-formula> and <inline-formula><math xmlns="http://www.w3.org/1998/Math/MathML" display="inline"><semantics><mrow><mo>Δ</mo><mi>V</mi></mrow></semantics></math></inline-formula> of P-Bit device variations by rederiving the weight matrix, enabling them to compute as ideal identical P-Bits without the need for weights retraining. Accurately extracting the <inline-formula><math xmlns="http://www.w3.org/1998/Math/MathML" display="inline"><semantics><mi>α</mi></semantics></math></inline-formula> and <inline-formula><math xmlns="http://www.w3.org/1998/Math/MathML" display="inline"><semantics><mrow><mo>Δ</mo><mi>V</mi></mrow></semantics></math></inline-formula> simultaneously from a large P-Bit array which is prerequisite for the weight compensation method is a crucial and challenging task. To solve this obstacle, we present the novel automatic variation extraction algorithm which can extract device variations of each P-Bit in a large array based on Boltzmann machine learning. In order for the accurate extraction of variations from an extendable P-Bit array, an Ising Hamiltonian based on a 3D ferromagnetic model is constructed, achieving precise and scalable array variation extraction. The proposed Automatic Extraction and Compensation algorithm is utilized to solve both 16-city traveling salesman problem (TSP) and 21-bit integer factorization on a large P-Bit array with variation, demonstrating its accuracy, transferability, and scalability.https://www.mdpi.com/2072-666X/16/2/133spintronicsmagnetic tunneling junctionIsing computationBoltzmann machine trainingtraveling salesman probleminteger factorization
spellingShingle Bolin Zhang
Yu Liu
Tianqi Gao
Jialiang Yin
Zhenyu Guan
Deming Zhang
Lang Zeng
Automatic Extraction and Compensation of P-Bit Device Variations in Large Array Utilizing Boltzmann Machine Training
Micromachines
spintronics
magnetic tunneling junction
Ising computation
Boltzmann machine training
traveling salesman problem
integer factorization
title Automatic Extraction and Compensation of P-Bit Device Variations in Large Array Utilizing Boltzmann Machine Training
title_full Automatic Extraction and Compensation of P-Bit Device Variations in Large Array Utilizing Boltzmann Machine Training
title_fullStr Automatic Extraction and Compensation of P-Bit Device Variations in Large Array Utilizing Boltzmann Machine Training
title_full_unstemmed Automatic Extraction and Compensation of P-Bit Device Variations in Large Array Utilizing Boltzmann Machine Training
title_short Automatic Extraction and Compensation of P-Bit Device Variations in Large Array Utilizing Boltzmann Machine Training
title_sort automatic extraction and compensation of p bit device variations in large array utilizing boltzmann machine training
topic spintronics
magnetic tunneling junction
Ising computation
Boltzmann machine training
traveling salesman problem
integer factorization
url https://www.mdpi.com/2072-666X/16/2/133
work_keys_str_mv AT bolinzhang automaticextractionandcompensationofpbitdevicevariationsinlargearrayutilizingboltzmannmachinetraining
AT yuliu automaticextractionandcompensationofpbitdevicevariationsinlargearrayutilizingboltzmannmachinetraining
AT tianqigao automaticextractionandcompensationofpbitdevicevariationsinlargearrayutilizingboltzmannmachinetraining
AT jialiangyin automaticextractionandcompensationofpbitdevicevariationsinlargearrayutilizingboltzmannmachinetraining
AT zhenyuguan automaticextractionandcompensationofpbitdevicevariationsinlargearrayutilizingboltzmannmachinetraining
AT demingzhang automaticextractionandcompensationofpbitdevicevariationsinlargearrayutilizingboltzmannmachinetraining
AT langzeng automaticextractionandcompensationofpbitdevicevariationsinlargearrayutilizingboltzmannmachinetraining