Patent · US Active

Representing a neural network utilizing paths within the network to improve a performance of the neural network

US11972354B2 · kind B2 · utility

0Cited by

1References

25Claims

0Family size

Assignee

NVIDIA Corporation · US

Inventors

Alexander Keller · Berlin, DE
Gonçalo Filipe Torcato Mordido · Potsdam, DE
Noah Gamboa · Menlo Park, US
Matthijs Jules Van Keirsbilck · Berlin, DE

Key dates

Filing date	Feb 15, 2022
Grant date	Apr 30, 2024
Priority date	—
Expiry date	May 15, 2042

Classification

Technology area (CPC G)Physics
CPC primaryG06N3/092
WIPO fieldComputer technology
WIPO sectorElectrical engineering

Abstract

Artificial neural networks (ANNs) are computing systems that imitate a human brain by learning to perform tasks by considering examples. By representing an artificial neural network utilizing individual paths each connecting an input of the ANN to an output of the ANN, a complexity of the ANN may be reduced, and the ANN may be trained and implemented in a much faster manner when compared to an implementation using fully connected ANN graphs.

Source: USPTO / EPO open patent data. Objective bibliographic and citation counts.