SwePub
Sök i LIBRIS databas

  Extended search

id:"swepub:oai:DiVA.org:kth-310412"
 

Search: id:"swepub:oai:DiVA.org:kth-310412" > StreamBrain :

  • 1 of 1
  • Previous record
  • Next record
  •    To hitlist
  • Podobas, ArturKTH,Beräkningsvetenskap och beräkningsteknik (CST) (author)

StreamBrain : An HPC Framework for Brain-like Neural Networks on CPUs, GPUs and FPGAs

  • Article/chapterEnglish2021

Publisher, publication year, extent ...

  • 2021-06-21
  • New York, NY, USA :Association for Computing Machinery (ACM),2021
  • printrdacarrier

Numbers

  • LIBRIS-ID:oai:DiVA.org:kth-310412
  • https://urn.kb.se/resolve?urn=urn:nbn:se:kth:diva-310412URI
  • https://doi.org/10.1145/3468044.3468052DOI

Supplementary language notes

  • Language:English
  • Summary in:English

Part of subdatabase

Classification

  • Subject category:ref swepub-contenttype
  • Subject category:kon swepub-publicationtype

Notes

  • Part of proceedings ISBN: 978-1-4503-8549-7QC 20220331
  • The modern deep learning method based on backpropagation has surged in popularity and has been used in multiple domains and application areas. At the same time, there are other - less-known - machine learning algorithms with a mature and solid theoretical foundation whose performance remains unexplored. One such example is the brain-like Bayesian Confidence Propagation Neural Network (BCPNN). In this paper, we introduce StreamBrain - a framework that allows neural networks based on BCPNN to be practically deployed in High-Performance Computing systems. StreamBrain is a domain-specific language (DSL), similar in concept to existing machine learning (ML) frameworks, and supports backends for CPUs, GPUs, and even FPGAs. We empirically demonstrate that StreamBrain can train the well-known ML benchmark dataset MNIST within seconds, and we are the first to demonstrate BCPNN on STL-10 size networks. We also show how StreamBrain can be used to train with custom floating-point formats and illustrate the impact of using different bfloat variations on BCPNN using FPGAs.

Subject headings and genre

Added entries (persons, corporate bodies, meetings, titles ...)

  • Svedin, MartinKTH,Beräkningsvetenskap och beräkningsteknik (CST)(Swepub:kth)u15sd73k (author)
  • Chien, Wei DerKTH,Beräkningsvetenskap och beräkningsteknik (CST)(Swepub:kth)u1qkq46g (author)
  • Peng, Ivy B.Lawrence Livermore National Laboratory, CA, USA (author)
  • Ravichandran, Naresh BalajiKTH,Beräkningsvetenskap och beräkningsteknik (CST)(Swepub:kth)u1d489nh (author)
  • Herman, Pawel,1979-KTH,Beräkningsvetenskap och beräkningsteknik (CST)(Swepub:kth)u19pqm1e (author)
  • Lansner, Anders,Professor,1949-KTH,Beräkningsvetenskap och beräkningsteknik (CST),Stockholm University, Stockholm, Sweden(Swepub:kth)u12s8cr8 (author)
  • Markidis, StefanoKTH,Beräkningsvetenskap och beräkningsteknik (CST)(Swepub:kth)u1y5heyt (author)
  • KTHBeräkningsvetenskap och beräkningsteknik (CST) (creator_code:org_t)

Related titles

  • In:ACM International Conference Proceeding SeriesNew York, NY, USA : Association for Computing Machinery (ACM)

Internet link

To the university's database

  • 1 of 1
  • Previous record
  • Next record
  •    To hitlist

Search outside SwePub

Kungliga biblioteket hanterar dina personuppgifter i enlighet med EU:s dataskyddsförordning (2018), GDPR. Läs mer om hur det funkar här.
Så här hanterar KB dina uppgifter vid användning av denna tjänst.

 
pil uppåt Close

Copy and save the link in order to return to this view