Patents by Inventor Serdar SOZUBEK

Serdar SOZUBEK 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).

  • Reduction of parameters in fully connected layers of neural networks by low rank factorizations
    Patent number: 10896366
    Abstract: The present disclosure is drawn to the reduction of parameters in fully connected layers of neural networks. For a layer whose output is defined by y=Wx, where y?Rm is the output vector, x?Rn is the input vector, and W?Rm×n is a matrix of connection parameters, matrices Uij and Vij are defined and submatrices Wij are computed as the product of Uij and Vij, so that Wij=VijUij, and W is obtained by appending submatrices Wi,j.
    Type: Grant
    Filed: March 8, 2017
    Date of Patent: January 19, 2021
    Assignee: Huawei Technologies Co., Ltd.
    Inventors: Serdar Sozubek, Barnaby Dalton, Vanessa Courville, Graham Taylor
  • SYSTEM AND METHOD FOR CASCADED MAX POOLING IN NEURAL NETWORKS
    Publication number: 20200090023
    Abstract: A method for performing size K×K max pooling with stride S at a pooling layer of a convolutional neural network to downsample input data includes receiving input data, buffering the input data, applying a cascade of size 2×2 pooling stages to the buffered input data to generate downsampled output data, and outputting the downsampled output data to another layer of the convolutional neural network for further processing.
    Type: Application
    Filed: September 14, 2018
    Publication date: March 19, 2020
    Inventors: John Joseph, Serdar Sozubek
  • SYSTEM AND METHOD FOR CASCADED DYNAMIC MAX POOLING IN NEURAL NETWORKS
    Publication number: 20200090046
    Abstract: A method for performing size K×K max pooling with stride S at a max pooling layer of a convolutional neural network to downsample input data includes receiving input data, buffering the input data, applying a cascade of size 2×2 max pooling stages to the buffered input data to generate downsampled output data, wherein a stride value of each size 2×2 max pooling stage is determined dynamically in accordance with pooling parameters associated with the size 2×2 max pooling stage.
    Type: Application
    Filed: September 14, 2018
    Publication date: March 19, 2020
    Inventors: Serdar Sozubek, John Joseph
  • Reduction of parameters in fully connected layers of neural networks
    Patent number: 10509996
    Abstract: The present disclosure is drawn to the reduction of parameters in fully connected layers of neural networks. For a layer whose output is defined by y=Wx, where y is the output vector, x is the input vector, and W is a matrix of connection parameters, vectors uij and vij are defined and submatrices Wi,j are computed as the outer product of uij and vij, so that Wi,j=vij?uij, and W is obtained by appending submatrices Wi,j.
    Type: Grant
    Filed: September 7, 2016
    Date of Patent: December 17, 2019
    Assignee: Huawei Technologies Co., Ltd.
    Inventors: Barnaby Dalton, Serdar Sozubek, Manuel Saldana, Vanessa Courville
  • REDUCTION OF PARAMETERS IN FULLY CONNECTED LAYERS OF NEURAL NETWORKS BY LOW RANK FACTORIZATIONS
    Publication number: 20170337465
    Abstract: The present disclosure is drawn to the reduction of parameters in fully connected layers of neural networks. For a layer whose output is defined by y=Wx, where y ? Rm is the output vector, x ? Rn is the input vector, and W ? Rmxn is a matrix of connection parameters, matrices Uij and Vij are defined and submatrices Wij are computed as the product of Uij and Vij, so that Wij=VijUij, and W is obtained by appending submatrices Wi,j.
    Type: Application
    Filed: March 8, 2017
    Publication date: November 23, 2017
    Inventors: Serdar SOZUBEK, Barnaby DALTON, Vanessa COURVILLE, Graham TAYLOR
  • REDUCTION OF PARAMETERS IN FULLY CONNECTED LAYERS OF NEURAL NETWORKS
    Publication number: 20170337463
    Abstract: The present disclosure is drawn to the reduction of parameters in fully connected layers of neural networks. For a layer whose output is defined by y=Wx, where y is the output vector, x is the input vector, and W is a matrix of connection parameters, vectors uij and ij are defined and submatrices Wi,j are computed as the outer product of uij and ij, so that Wi,j=ijuij, and W is obtained by appending submatrices Wi,j.
    Type: Application
    Filed: September 7, 2016
    Publication date: November 23, 2017
    Inventors: Barnaby DALTON, Serdar SOZUBEK, Manuel SALDANA, Vanessa COURVILLE