Title:
効率的な行列乗算のためのシステムおよび方法
Document Type and Number:
Japanese Patent JP7130766
Kind Code:
B2
Abstract:
Disclosed are systems and methods for performing efficient vector-matrix multiplication using a sparsely-connected conductance matrix and analog mixed signal (AMS) techniques. Metal electrodes are sparsely connected using coaxial nanowires. Each electrode can be used as an input/output node or neuron in a neural network layer. Neural network synapses are created by random connections provided by coaxial nanowires. A subset of the metal electrodes can be used to receive a vector of input voltages and the complementary subset of the metal electrodes can be used to read output currents. The output currents are the result of vector-matrix multiplication of the vector of input voltages with the sparsely-connected matrix of conductances.
Inventors:
Kendal Jack David
Application Number:
JP2020551826A
Publication Date:
September 05, 2022
Filing Date:
April 05, 2019
Export Citation:
Assignee:
RAIN NEUROMORPHICS INC.
International Classes:
G06G7/60; G06F17/16; G06N3/063
Domestic Patent References:
| JP2017527000A | ||||
| JP2003263624A | ||||
| JP2005182785A |
Foreign References:
| US20170228345 |
Attorney, Agent or Firm:
Patent Attorney Corporation Meisei International Patent Office
Previous Patent: Head mounted display and head mounted display system
Next Patent: Event monitoring method and apparatus
Next Patent: Event monitoring method and apparatus