Systems and methods for efficient matrix multiplication
AssigneeRain Neuromorphics Inc.
InventorJack David Kendall
Life of this patent
Estimating the term…
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.
Claims19
A system of sparse vector-matrix multiplication comprising:
- a circuit layer;
- a plurality of electrodes on the circuit layer;
- and
- a mesh on and sparsely coupled to the plurality of electrodes, the mesh comprising a plurality of nanowires forming an irregular network, wherein the circuit layer is configured to:
- receive a first plurality of input signals;
- write a second plurality of input signals on an input set of the plurality of electrodes;
- read from an output set of the plurality of electrodes a first plurality of output signals;
- and
- output a second plurality of output signals.
A method of providing a sparse vector-matrix multiplication system, comprising:
- providing a circuit layer;
- providing a plurality of electrodes on the circuit layer;
- providing a plurality of coaxial nanowires forming an irregular network on the plurality of electrodes, the plurality of coaxial nanowires being arranged in a mesh;
- and
- sparsely coupling the plurality of coaxial nanowires to the plurality of electrodes.
A method of performing a sparse vector-matrix multiplication, comprising:
- receiving at an input set of a plurality of electrodes a plurality of input signals from a circuit layer, the plurality of electrodes being sparsely connected to a plurality of coaxial nanowires on the plurality of electrodes, the plurality of coaxial nanowires forming an irregular network and being arranged in a mesh;
- and
- reading from an output set of the plurality of electrodes a plurality of output signals.
A method of performing vector-matrix multiplication, comprising:
- receiving at an input set of a plurality of electrodes a plurality of input signals from a circuit layer, the plurality of electrodes being sparsely connected to a plurality of coaxial nanowires on the plurality of electrodes, wherein the coaxial nanowires comprise a metal core wrapped in two-terminal non-volatile memory (NVM) material, wherein the NVM material comprises one or more of a voltage-controlled resistance, memristor, phase-change material (PCM), and resistive random-access-memory (ReRAM) material, wherein voltage-controlled resistances are at intersections of the plurality of the electrodes and the plurality of coaxial nanowires;
- selectively providing training voltages to at least a portion of the plurality of the electrodes;
- reading from an output set of the plurality of electrodes a plurality of output signals;
- and
- receiving a training signal indicating which electrodes in the plurality of the electrodes are to be applied the training voltages.
Dependencies are inferred from the claim text.