TY  - CONF
EP  - 2513
A1  - Hamid, N.H.
A1  - Murray, A.F.
A1  - Laurenson, D.
A1  - Roy, S.
A1  - Cheng, B.
UR  - https://www.scopus.com/inward/record.uri?eid=2-s2.0-42949129603&doi=10.1109%2fISCAS.2005.1465136&partnerID=40&md5=20e8c5d6943a9041fa79b586befa3b2a
SN  - 02714310
Y1  - 2005///
ID  - scholars2
SP  - 2510
TI  - Probabilistic computing with future deep sub-micrometer devices : A modelling approach
KW  - MOS technology; Mosfet model; MOSFETs; N-channel; Probabilistic computing; Restricted boltzmann machine; Stochastic behavior; Submicrometer
KW  -  Electric load management; Frequency multiplying circuits; MATLAB; Micrometers; MOSFET devices; Random processes
KW  -  Stochastic models
N1  - cited By 3; Conference of IEEE International Symposium on Circuits and Systems 2005, ISCAS 2005 ; Conference Date: 23 May 2005 Through 26 May 2005; Conference Code:77312
N2  - An approach is described that investigates the potential of probabilistic "neural" architectures for computation with Deep Sub-Micrometer (DSM) MOSFETs. Initially, noisy MOSFET models are based upon those for a 0.35μm MOS technology with an exaggerated 1/f characteristic. We explore the manifestation of the 1/f characteristic at the output of 2-quadrant multiplier when the key n-channel MOSFETs are replaced by "noisy" MOSFETs. The stochastic behavior of this noisy multiplier has been mapped on to a software (Matlab) model of a Continuous Restricted Boltzmann Machine (CRBM) - an analogue-input stochastic computing structure. Simulation of this DSM CRBM implementation shows little degradation from that of a "perfect" CRBM. This paper thus introduces a methodology for a form of "technology-downstreaming" and highlights the potential of probabilistic architectures for DSM computation. © 2005 IEEE.
AV  - none
CY  - Kobe
ER  -