Configurable p-Neurons Using Modular p-Bits

Fuente: arXiv
Enregistré dans:
Détails bibliographiques
Auteurs principaux: Bunaiyan, Saleh, Alsharif, Mohammad, Abdelrahman, Abdelrahman S., ElSawy, Hesham, Cheema, Suraj S., Fahmy, Suhaib A., Camsari, Kerem Y., Al-Dirini, Feras
Format: Preprint
Publié: 2026
Sujets:
Accès en ligne:
Tags: Ajouter un tag
Pas de tags, Soyez le premier à ajouter un tag!
_version_ 1866917224780922880
author Bunaiyan, Saleh
Alsharif, Mohammad
Abdelrahman, Abdelrahman S.
ElSawy, Hesham
Cheema, Suraj S.
Fahmy, Suhaib A.
Camsari, Kerem Y.
Al-Dirini, Feras
author_facet Bunaiyan, Saleh
Alsharif, Mohammad
Abdelrahman, Abdelrahman S.
ElSawy, Hesham
Cheema, Suraj S.
Fahmy, Suhaib A.
Camsari, Kerem Y.
Al-Dirini, Feras
contents Probabilistic bits (p-bits) have recently been employed in neural networks (NNs) as stochastic neurons with sigmoidal probabilistic activation functions. Nonetheless, there remain a wealth of other probabilistic activation functions that are yet to be explored. Here we re-engineer the p-bit by decoupling its stochastic signal path from its input data path, giving rise to a modular p-bit that enables the realization of probabilistic neurons (p-neurons) with a range of configurable probabilistic activation functions, including a probabilistic version of the widely used Logistic Sigmoid, Tanh and Rectified Linear Unit (ReLU) activation functions. We present spintronic (CMOS + sMTJ) designs that show wide and tunable probabilistic ranges of operation. Finally, we experimentally implement digital-CMOS versions on an FPGA, with stochastic unit sharing, and demonstrate an order of magnitude (10x) saving in required hardware resources compared to conventional digital p-bit implementations.
format Preprint
id arxiv_https___arxiv_org_abs_2601_18943
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle Configurable p-Neurons Using Modular p-Bits
Bunaiyan, Saleh
Alsharif, Mohammad
Abdelrahman, Abdelrahman S.
ElSawy, Hesham
Cheema, Suraj S.
Fahmy, Suhaib A.
Camsari, Kerem Y.
Al-Dirini, Feras
Emerging Technologies
Disordered Systems and Neural Networks
Artificial Intelligence
Hardware Architecture
Machine Learning
Probabilistic bits (p-bits) have recently been employed in neural networks (NNs) as stochastic neurons with sigmoidal probabilistic activation functions. Nonetheless, there remain a wealth of other probabilistic activation functions that are yet to be explored. Here we re-engineer the p-bit by decoupling its stochastic signal path from its input data path, giving rise to a modular p-bit that enables the realization of probabilistic neurons (p-neurons) with a range of configurable probabilistic activation functions, including a probabilistic version of the widely used Logistic Sigmoid, Tanh and Rectified Linear Unit (ReLU) activation functions. We present spintronic (CMOS + sMTJ) designs that show wide and tunable probabilistic ranges of operation. Finally, we experimentally implement digital-CMOS versions on an FPGA, with stochastic unit sharing, and demonstrate an order of magnitude (10x) saving in required hardware resources compared to conventional digital p-bit implementations.
title Configurable p-Neurons Using Modular p-Bits
topic Emerging Technologies
Disordered Systems and Neural Networks
Artificial Intelligence
Hardware Architecture
Machine Learning
url https://arxiv.org/abs/2601.18943