Enregistré dans:
Détails bibliographiques
Auteurs principaux: Patel, Saavan, Canoza, Philip, Datar, Adhiraj, Lu, Steven, Garg, Chirag, Salahuddin, Sayeef
Format: Preprint
Publié: 2024
Sujets:
Accès en ligne:https://arxiv.org/abs/2409.10325
Tags: Ajouter un tag
Pas de tags, Soyez le premier à ajouter un tag!
_version_ 1866910605408993280
author Patel, Saavan
Canoza, Philip
Datar, Adhiraj
Lu, Steven
Garg, Chirag
Salahuddin, Sayeef
author_facet Patel, Saavan
Canoza, Philip
Datar, Adhiraj
Lu, Steven
Garg, Chirag
Salahuddin, Sayeef
contents New computing paradigms are required to solve the most challenging computational problems where no exact polynomial time solution exists.Probabilistic Ising Accelerators has gained promise on these problems with the ability to model complex probability distributions and find ground states of intractable problems. In this context, we have demonstrated the Parallel Asynchronous Stochastic Sampler (PASS), the first fully on-chip integrated, asynchronous, probabilistic accelerator that takes advantage of the intrinsic fine-grained parallelism of the Ising Model and built in state of the art 14nm CMOS FinFET technology. We have demonstrated broad applicability of this accelerator on problems ranging from Combinatorial Optimization, Neural Simulation, to Machine Learning along with up to $23,000$x energy to solution improvement compared to CPUs on probabilistic problems.
format Preprint
id arxiv_https___arxiv_org_abs_2409_10325
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle PASS: An Asynchronous Probabilistic Processor for Next Generation Intelligence
Patel, Saavan
Canoza, Philip
Datar, Adhiraj
Lu, Steven
Garg, Chirag
Salahuddin, Sayeef
Distributed, Parallel, and Cluster Computing
Hardware Architecture
Emerging Technologies
Data Analysis, Statistics and Probability
New computing paradigms are required to solve the most challenging computational problems where no exact polynomial time solution exists.Probabilistic Ising Accelerators has gained promise on these problems with the ability to model complex probability distributions and find ground states of intractable problems. In this context, we have demonstrated the Parallel Asynchronous Stochastic Sampler (PASS), the first fully on-chip integrated, asynchronous, probabilistic accelerator that takes advantage of the intrinsic fine-grained parallelism of the Ising Model and built in state of the art 14nm CMOS FinFET technology. We have demonstrated broad applicability of this accelerator on problems ranging from Combinatorial Optimization, Neural Simulation, to Machine Learning along with up to $23,000$x energy to solution improvement compared to CPUs on probabilistic problems.
title PASS: An Asynchronous Probabilistic Processor for Next Generation Intelligence
topic Distributed, Parallel, and Cluster Computing
Hardware Architecture
Emerging Technologies
Data Analysis, Statistics and Probability
url https://arxiv.org/abs/2409.10325