Zobrazeno 1 - 10
of 464
pro vyhledávání: '"Peter Y K Cheung"'
Publikováno v:
IEEE Access, Vol 12, Pp 54272-54284 (2024)
With the rapid advancement of quantitative trading technology, the demand for low-latency in Level 2 Deep Market Quote (L2DMQ) Decoding is ever-increasing. The L2DMQ decoder faces increasingly significant challenges in terms of bandwidth and performa
Externí odkaz:
https://doaj.org/article/5d0ed10f90eb4092b812ad747e65d39b
Autor:
Michelangelo Bin, Peter Y K Cheung, Emanuele Crisostomi, Pietro Ferraro, Hugo Lhachemi, Roderick Murray-Smith, Connor Myant, Thomas Parisini, Robert Shorten, Sebastian Stein, Lewi Stone
Publikováno v:
PLoS Computational Biology, Vol 17, Iss 1, p e1008604 (2021)
COVID-19 abatement strategies have risks and uncertainties which could lead to repeating waves of infection. We show-as proof of concept grounded on rigorous mathematical evidence-that periodic, high-frequency alternation of into, and out-of, lockdow
Externí odkaz:
https://doaj.org/article/51a8e062cb0b4380a55ba90c59416a08
Autor:
Erwei Wang, Marie Auffret, Georgios-Ilias Stavrou, Peter Y. K. Cheung, George A. Constantinides, Mohamed S. Abdelfattah, James J. Davis
FPGA-specific DNN architectures using the native LUTs as independently trainable inference operators have been shown to achieve favorable area-accuracy and energy-accuracy tradeoffs. The first work in this area, LUTNet, exhibited state-of-the-art per
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_dedup___::b4cf6d3ba35d01c6dcd5ea66bdfda710
http://hdl.handle.net/10044/1/103101
http://hdl.handle.net/10044/1/103101
Autor:
Maurizio Pierini, Eric A. Moreno, Erwei Wang, Bartlomiej Borzyszkowski, Sioni Summers, Vladimir Loncar, Peter Y. K. Cheung, Thea Klaeboe Aarrestad, Zhiqiang Que, Umar Marikar, Jennifer Ngadiuba, Wayne Luk, Hamza Javed
Publikováno v:
ASAP
32nd IEEE International Conference on Application-specific Systems, Architectures and Processors (ASAP)
32nd IEEE International Conference on Application-specific Systems, Architectures and Processors (ASAP)
This paper presents novel reconfigurable architectures for reducing the latency of recurrent neural networks (RNNs) that are used for detecting gravitational waves. Gravitational interferometers such as the LIGO detectors capture cosmic events such a
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_dedup___::7e76574c7a0952d39cbfa86492383378
https://doi.org/10.1109/asap52443.2021.00025
https://doi.org/10.1109/asap52443.2021.00025
Autor:
Erwei Wang, Piotr Zieliński, Daniele Moro, George A. Constantinides, Jia Jie Lim, Claudionor Coelho, James J. Davis, Peter Y. K. Cheung, Satrajit Chatterjee
Publikováno v:
International Workshop on Embedded and Mobile Deep Learning (EDML)
Workshop on Binary Networks for Computer Vision
EMDL@MobiSys
Workshop on Binary Networks for Computer Vision
EMDL@MobiSys
The ever-growing computational demands of increasingly complex machine learning models frequently necessitate the use of powerful cloud-based infrastructure for their training. Binary neural networks are known to be promising candidates for on-device
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_dedup___::f23e8ff935bfe5b252a0d913cde9e470
http://arxiv.org/abs/2102.04270
http://arxiv.org/abs/2102.04270
Autor:
Erwei Wang, James J. Davis, Georgios-Ilias Stavrou, Peter Y. K. Cheung, George A. Constantinides, Mohamed Abdelfattah
Publikováno v:
ACM/SIGDA International Symposium on Field-Programmable Gate Arrays
FPGA-specific DNN architectures using the native LUTs as independently trainable inference operators have been shown to achieve favorable area-accuracy and energy-accuracy tradeoffs. The first work in this area, LUTNet, exhibited state-of-the-art per
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_dedup___::87fc209d9404fa50833ae69bb517008d
Autor:
Roderick Murray-Smith, Pietro Ferraro, Lewi Stone, Emanuele Crisostomi, Thomas Parisini, Michelangelo Bin, Sebastian Stein, Connor Myant, Hugo Lhachemi, Robert Shorten, Peter Y. K. Cheung
Publikováno v:
PLoS Computational Biology
PLoS Computational Biology, Public Library of Science, 2021, 17 (1), pp.e1008604. ⟨10.1371/journal.pcbi.1008604⟩
PLOS Computational Biology
PLoS Computational Biology, Vol 17, Iss 1, p e1008604 (2021)
PLoS Computational Biology, Public Library of Science, 2021, 17 (1), pp.e1008604. ⟨10.1371/journal.pcbi.1008604⟩
PLOS Computational Biology
PLoS Computational Biology, Vol 17, Iss 1, p e1008604 (2021)
COVID-19 abatement strategies have risks and uncertainties which could lead to repeating waves of infection. We show—as proof of concept grounded on rigorous mathematical evidence—that periodic, high-frequency alternation of into, and out-of, loc
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_dedup___::775d9fa90c157d40d1d9b0d700bffdc2
https://hdl.handle.net/11368/2993761
https://hdl.handle.net/11368/2993761
Research has shown that deep neural networks contain significant redundancy, and thus that high classification accuracy can be achieved even when weights and activations are quantized down to binary values. Network binarization on FPGAs greatly incre
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_dedup___::2023f3d7086ca2b489c52afd8a81ccc3
http://hdl.handle.net/10044/1/80013
http://hdl.handle.net/10044/1/80013
Autor:
Peter Y. K. Cheung
Publikováno v:
Proceedings of the 2020 ACM/SIGDA International Symposium on Field-Programmable Gate Arrays.
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_________::3a9e84f58b8c0e7fdbd823130c701b36
https://doi.org/10.1145/3384500
https://doi.org/10.1145/3384500
Publikováno v:
International Journal of Reconfigurable Computing, Vol 2010 (2010)
This paper proposes a benchmarking methodology for characterising the power consumption of the fine-grain fabric in reconfigurable architectures. This methodology is part of the GroundHog 2009 power benchmarking suite. It covers active and inactive p
Externí odkaz:
https://doaj.org/article/e92b1116585440e5aab6dbcb29b45366