Processing

Please wait...

Settings

Settings

Goto Application

1. WO2020223206 - SUPERCONDUCTING PARAMETRIC AMPLIFIER NEURAL NETWORK

Publication Number WO/2020/223206
Publication Date 05.11.2020
International Application No. PCT/US2020/030220
International Filing Date 28.04.2020
IPC
G06N 3/04 2006.01
GPHYSICS
06COMPUTING; CALCULATING OR COUNTING
NCOMPUTER SYSTEMS BASED ON SPECIFIC COMPUTATIONAL MODELS
3Computer systems based on biological models
02using neural network models
04Architecture, e.g. interconnection topology
G06N 3/06 2006.01
GPHYSICS
06COMPUTING; CALCULATING OR COUNTING
NCOMPUTER SYSTEMS BASED ON SPECIFIC COMPUTATIONAL MODELS
3Computer systems based on biological models
02using neural network models
06Physical realisation, i.e. hardware implementation of neural networks, neurons or parts of neurons
G06N 3/063 2006.01
GPHYSICS
06COMPUTING; CALCULATING OR COUNTING
NCOMPUTER SYSTEMS BASED ON SPECIFIC COMPUTATIONAL MODELS
3Computer systems based on biological models
02using neural network models
06Physical realisation, i.e. hardware implementation of neural networks, neurons or parts of neurons
063using electronic means
G06N 3/02 2006.01
GPHYSICS
06COMPUTING; CALCULATING OR COUNTING
NCOMPUTER SYSTEMS BASED ON SPECIFIC COMPUTATIONAL MODELS
3Computer systems based on biological models
02using neural network models
CPC
G06N 10/00
GPHYSICS
06COMPUTING; CALCULATING; COUNTING
NCOMPUTER SYSTEMS BASED ON SPECIFIC COMPUTATIONAL MODELS
10Quantum computers, i.e. computer systems based on quantum-mechanical phenomena
G06N 3/04
GPHYSICS
06COMPUTING; CALCULATING; COUNTING
NCOMPUTER SYSTEMS BASED ON SPECIFIC COMPUTATIONAL MODELS
3Computer systems based on biological models
02using neural network models
04Architectures, e.g. interconnection topology
G06N 3/063
GPHYSICS
06COMPUTING; CALCULATING; COUNTING
NCOMPUTER SYSTEMS BASED ON SPECIFIC COMPUTATIONAL MODELS
3Computer systems based on biological models
02using neural network models
06Physical realisation, i.e. hardware implementation of neural networks, neurons or parts of neurons
063using electronic means
H03K 19/1954
HELECTRICITY
03BASIC ELECTRONIC CIRCUITRY
KPULSE TECHNIQUE
19Logic circuits, i.e. having at least two inputs acting on one output
02using specified components
195using superconductive devices
1954with injection of the control current
Applicants
  • MASSACHUSETTS INSTITUTE OF TECHNOLOGY [US]/[US]
Inventors
  • WYNN, Alexander
Agents
  • LANGE, Kristoffer, W.
  • DALY, Christopher, S.
  • BLAU, David, E.
  • KIM, Do Te
  • DIMOV, Kiril, O.
  • ROBINSON, Kermit
  • MOOSEY, Anthony, T.
  • DURKEE, Paul, D.
  • CROWLEY, Judith, C.
  • MOFFORD, Donald, F.
  • DOWNING, Marianne, M.
  • DUBUC, Marisa, J.
Priority Data
62/839,88529.04.2019US
Publication Language English (EN)
Filing Language English (EN)
Designated States
Title
(EN) SUPERCONDUCTING PARAMETRIC AMPLIFIER NEURAL NETWORK
(FR) RÉSEAU NEURONAL D'AMPLIFICATEUR PARAMÉTRIQUE SUPRACONDUCTEUR
Abstract
(EN)
In some embodiments, a superconducting parametric amplification neural network (SPANN) includes neurons that operate in the analog domain, and a fanout network coupling the neurons that operates in the digital domain. Each neuron is provided one or more input currents having a resolution of several bits. The neuron weights the currents, sums the weighted currents with an optional bias or threshold current, then applies a nonlinear activation function to the result. The nonlinear function is implemented using a quantum flux parametron (QFP), thereby simultaneously amplifying and digitizing the output current signal. The digitized output of some or all neurons in each layer is provided to the next layer using a fanout network that operates to preserve the digital information held in the current.
(FR)
Dans certains modes de réalisation, un réseau neuronal d'amplification paramétrique supraconducteur (SPANN) comprend des neurones qui fonctionnent dans le domaine analogique, et un réseau de sortance couplant les neurones qui fonctionne dans le domaine numérique. Chaque neurone comprend un ou plusieurs courants d'entrée ayant une résolution de plusieurs bits. Le neurone pondère les courants, additionne les courants pondérés avec un courant de polarisation ou de seuil facultatif, puis applique une fonction d'activation non linéaire au résultat. La fonction non linéaire est mise en œuvre à l'aide d'un paramètre de flux quantique (QFP), ce qui permet d'amplifier et de numériser simultanément le signal de courant de sortie. La sortie numérisée de certains ou de tous les neurones dans chaque couche est fournie à la couche suivante à l'aide d'un réseau de sortance qui fonctionne pour préserver les informations numériques conservées dans le courant.
Latest bibliographic data on file with the International Bureau