Zobrazeno 1 - 10
of 41
pro vyhledávání: '"Chiu Pi-Feng"'
In cloud and edge computing models, it is important that compute devices at the edge be as power efficient as possible. Long short-term memory (LSTM) neural networks have been widely used for natural language processing, time series prediction and ma
Externí odkaz:
http://arxiv.org/abs/2002.10636
Autor:
Tung Thanh Hoang, Wen Ma, Martin Lueker-Boden, Chiu Pi-Feng, Won Ho Choi, Zvonimir Z. Bandic, Gertjan Hemink
Publikováno v:
ISCAS
An SLC NAND array based in-flash computing core is proposed for enabling vector-matrix multiplications in binarized neural network (BNN) and binary weight network (BWN). Two SLC NAND floating gate (FG) cells in the same string store complementary dat
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_________::3a73a28549a0ea799f2967fa29c7f1a9
https://doi.org/10.1109/iscas45731.2020.9180920
https://doi.org/10.1109/iscas45731.2020.9180920
Autor:
Rick Galbraith, Daniel Bedau, Dirk J. Wouters, T. Hennen, J. A. J. Rupp, Chiu Pi-Feng, Won Ho Choi, Rainer Waser, Jonas A. Goode, Martin Lueker-Boden, Wen Ma
Publikováno v:
ISCAS
In this work, we use a Cr-doped V2O3 based Mott oscillator circuit to build a reservoir computing system that has much smaller model size than an equivalent LSTM. In contrast to an LSTM, our reservoir computing system can be trained very efficiently
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_________::ac0faa76a573839024c03faa7bfa2a00
https://doi.org/10.1109/iscas45731.2020.9181105
https://doi.org/10.1109/iscas45731.2020.9181105
Akademický článek
Tento výsledek nelze pro nepřihlášené uživatele zobrazit.
K zobrazení výsledku je třeba se přihlásit.
K zobrazení výsledku je třeba se přihlásit.
Publikováno v:
ICRC
In cloud and edge computing models, it is important that compute devices at the edge be as power efficient as possible. Long short-term memory (LSTM) neural networks have been widely used for natural language processing, time series prediction and ma
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_dedup___::afeeda02d88674c24af13a437e999a49
https://doi.org/10.1109/icrc.2019.8914713
https://doi.org/10.1109/icrc.2019.8914713
Publikováno v:
ISCAS
Binarized Neural Networks (BNN) significantly reduce computational complexity and relax memory requirements with binarized weights and activations. We propose a differential crosspoint (DX) memristor array for enabling parallel multiply-and-accumulat
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_________::8b3c1fccfd36e76a56512914005c4167
https://doi.org/10.1109/iscas.2019.8702128
https://doi.org/10.1109/iscas.2019.8702128
Publikováno v:
ICMLA
The noise effect during training and inference for deep artificial neural network hardware acceleration is analyzed in this paper. The noise effect is extremely important when designing hardware for machine learning due to non-ideal devices and circu
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_________::342807f5a62250a512dff3c6ea7c4b88
https://doi.org/10.1109/icmla.2018.00127
https://doi.org/10.1109/icmla.2018.00127
Autor:
Celio, Christopher1 (AUTHOR), Chiu, Pi-Feng1 (AUTHOR), Asanovic, Krste1 (AUTHOR), Nikolic, Borivoje1 (AUTHOR), Patterson, David1 (AUTHOR)
Publikováno v:
IEEE Micro. Mar/Apr2019, Vol. 39 Issue 2, p52-60. 9p.
To continue reducing voltage in scaled technologies, both circuit and architecture-level resiliency techniques are needed to tolerate process-induced defects, variation, and aging in SRAM cells. Many different resiliency schemes have been proposed an
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=dedup_wf_001::1b5e1c0b4d0bf0917835bfda1dc67a2b
https://kluedo.ub.uni-kl.de/frontdoor/index/index/docId/4314
https://kluedo.ub.uni-kl.de/frontdoor/index/index/docId/4314
Akademický článek
Tento výsledek nelze pro nepřihlášené uživatele zobrazit.
K zobrazení výsledku je třeba se přihlásit.
K zobrazení výsledku je třeba se přihlásit.