Zobrazeno 1 - 6
of 6
pro vyhledávání: '"Junka Okazawa"'
Autor:
Akiyo Nomura, Megumi Ito, Sangbum Kim, Atsuya Okazaki, Masatoshi Ishii, Kohji Hosokawa, Wilfried Haensch, Junka Okazawa
Publikováno v:
Neural Processing Letters. 53:1741-1751
Neuromorphic computing using analog non-volatile memory (NVM) devices has been the subject of various studies due to its potential ability to achieve extremely low power consumption less than that of traditional von Neumann architecture. However, usi
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_________::206bddaab11decda0ab139993bd8fa96
https://doi.org/10.1007/s11063-019-10139-0
https://doi.org/10.1007/s11063-019-10139-0
Autor:
Scott C. Lewis, Atsuya Okazaki, Kohji Hosokawa, Junka Okazawa, Megumi Ito, Wanki Kim, Masatoshi Ishii, U. Shin, Wilfried Haensch, Matthew J. BrightSky, Malte J. Rasch, Akiyo Nomura, Sungchul Kim
Publikováno v:
2019 IEEE International Electron Devices Meeting (IEDM).
A fully silicon-integrated restricted Boltzmann machine (RBM) with event-driven contrastive divergence (eCD) algorithm is implemented using novel stochastic leaky integrate-and-fire (LIF) neuron circuits and 6-transistor/2- PCM-resistor (6T2R) unit c
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_________::502bd36e50ebdf0d2c6fd147176ef9f2
https://doi.org/10.1109/iedm19573.2019.8993466
https://doi.org/10.1109/iedm19573.2019.8993466
Training Large-Scale Spiking Neural Networks on Multi-core Neuromorphic System Using Backpropagation
Autor:
Akiyo Nomura, Atsuya Okazaki, Kohji Hosokawa, Sangbum Kim, Megumi Ito, Junka Okazawa, Masatoshi Ishii, Malte J. Rasch, Wilfried Haensch
Publikováno v:
Neural Information Processing ISBN: 9783030367176
ICONIP (3)
ICONIP (3)
Neuromorphic circuits with nonvolatile memory crossbar arrays can train and inference neural networks in a highly power-efficient manner, which can be a solution to overcome the von Neumann bottleneck. This paper proposes a scalable multi-core spikin
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_________::9c72ba001598e00ab64e513833821ac9
https://doi.org/10.1007/978-3-030-36718-3_16
https://doi.org/10.1007/978-3-030-36718-3_16
Autor:
Matthew J. BrightSky, Wanki Kim, Masatoshi Ishii, Atsuya Okazaki, Megumi Ito, Akiyo Nomura, Wilfried Haensch, Kohji Hosokawa, Junka Okazawa, Sangbum Kim
Publikováno v:
Communications in Computer and Information Science ISBN: 9783030368012
ICONIP (5)
ICONIP (5)
Phase change memory (PCM) is arguably one of the most promising non-volatile memories which can be used in neuromorphic applications. When we use PCM as analog synaptic elements, non-ideality impact should be carefully taken into account. In this pap
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_________::4147cc89a3906ad554b1af64ac32bb5a
https://doi.org/10.1007/978-3-030-36802-9_63
https://doi.org/10.1007/978-3-030-36802-9_63
Autor:
Masatoshi Ishii, Sangbum Kim, Kohji Hosokawa, Akiyo Nomura, Atsuya Okazaki, Junka Okazawa, Wilfried Haensch, Megumi Ito
Publikováno v:
Neural Information Processing ISBN: 9783030042387
ICONIP (7)
ICONIP (7)
In extremely energy-efficient neuromorphic computing using analog non-volatile memory (NVM) devices, device variability arises due to process variation and electro/thermo-dynamics of NVM devices, such as phase change memory and resistive-RAM. Thus, f
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_________::c7ced17c541fcd4ca032d5aea2d3d95a
https://doi.org/10.1007/978-3-030-04239-4_61
https://doi.org/10.1007/978-3-030-04239-4_61
Autor:
C. W. Yeh, Yu Zhu, Asit Kumar Ray, Mark D. Drapa, Chung H. Lam, Junka Okazawa, Scott C. Lewis, Sangbum Kim, Wanki Kim, Matt BrightSky, Tu-Shun Chen, Tony Perri, Robert L. Bruce, Jack Morrish, Huai-Yu Cheng, Chia-Jung Chen, Kohji Hosokawa, Wei-Chih Chien, Thomas M. Maffitt, Richard C. Jordan, Yutaka Nakamura, Hsiang-Lan Lung, Yung-Han Ho, H. Y. Ho, Christopher P. Miller, Jerry Heath
Publikováno v:
2016 IEEE 8th International Memory Workshop (IMW).
For the first time, by using a novel multiple individual bank sensing/writing and a memory bank interleave design, we demonstrate a double date rate 2 (DDR2) DRAM like interface phase-change memory (PCM) for M-type storage class memory applications .
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_________::bf2f81e529399561583fde9227489172
https://doi.org/10.1109/imw.2016.7493563
https://doi.org/10.1109/imw.2016.7493563