Patents by Inventor Simon Layton

Simon Layton has filed for patents to protect the following inventions. This listing includes patent applications that are pending as well as patents that have already been granted by the United States Patent and Trademark Office (USPTO).

  • Distributed weight update for backpropagation of a neural network
    Patent number: 12511543
    Abstract: Speed of training a neural network is improved by updating the weights of the neural network in parallel. In at least one embodiment, after back propagation, gradients are distributed to a plurality of processors, each of which calculate a portion of the updated weights of the neural network.
    Type: Grant
    Filed: November 5, 2019
    Date of Patent: December 30, 2025
    Assignee: NVIDIA Corporation
    Inventors: Sharan Chetlur, Natalia Gimelshein, Thor Mikal Johnsen, Simon Layton
  • DISTRIBUTED WEIGHT UPDATE FOR BACKPROPAGATION OF A NEURAL NETWORK
    Publication number: 20210133583
    Abstract: Speed of training a neural network is improved by updating the weights of the neural network in parallel. In at least one embodiment, after back propagation, gradients are distributed to a plurality of processors, each of which calculate a portion of the updated weights of the neural network.
    Type: Application
    Filed: November 5, 2019
    Publication date: May 6, 2021
    Inventors: Sharan Chetlur, Natalia Gimelshein, Thor Mikal Johnsen, Simon Layton