Zobrazeno 1 - 3
of 3
pro vyhledávání: '"Tianchu Ji"'
Publikováno v:
IEEE Micro. 39:17-25
In this article, we present Argus, an end-to-end framework for accelerating convolutional neural networks (CNNs) on field-programmable gate arrays (FPGAs) with minimum user effort. Argus uses state-of-the-art methods to auto-generate highly optimized
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_________::7cd6ca3a3dc82e175b4a928d0a5fe117
https://doi.org/10.1109/mm.2019.2930607
https://doi.org/10.1109/mm.2019.2930607
Autor:
Tianchu Ji, Shraddhan Jain, H. Andrew Schwartz, Michael Ferdman, Peter Milder, Niranjan Balasubramanian
Publikováno v:
ACL/IJCNLP (Findings)
How much information do NLP tasks really need from a transformer's attention mechanism at application-time (inference)? From recent work, we know that there is sparsity in transformers and that the floating-points within its computation can be discre
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_dedup___::eb117c10712375766bd293472d9365d3
https://doi.org/10.18653/v1/2021.findings-acl.363
https://doi.org/10.18653/v1/2021.findings-acl.363
Publikováno v:
FPL
To cope with the increasing demand and computational intensity of deep neural networks (DNNs), industry and academia have turned to accelerator technologies. In particular, FPGAs have been shown to provide a good balance between performance and energ
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_dedup___::7ec81267627aad148a40ebcb63992325
http://arxiv.org/abs/1807.04013
http://arxiv.org/abs/1807.04013
