Patents by Inventor Matthew Raja Khoury

Matthew Raja Khoury 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).

  • Processing matrix operations for rate limited systems
    Patent number: 11657262
    Abstract: Data to be processed includes vector element values of an input vector and matrix element values of a model matrix associated with a neural network model. A vector-matrix multiplication module receives a set of matrix element values for performing a vector-matrix multiplication operation. Processing the data includes computing a plurality of intermediate vectors based on element-wise vector multiplication between different subsets of the vector element values and different respective pre-processing vectors. The vector-matrix multiplication module is loaded with a core matrix, and the input vector is multiplied by the model matrix based on separately multiplying each of the intermediate vectors by the loaded core matrix.
    Type: Grant
    Filed: November 11, 2022
    Date of Patent: May 23, 2023
    Assignee: Lightelligence, Inc.
    Inventors: Matthew Raja Khoury, Rumen Rumenov Dangovski, Longwu Ou, Yichen Shen, Li Jing
  • Processing Matrix Operations for Rate Limited Systems
    Publication number: 20230077270
    Abstract: Data to be processed includes vector element values of an input vector and matrix element values of a model matrix associated with a neural network model. A vector-matrix multiplication module receives a set of matrix element values for performing a vector-matrix multiplication operation. Processing the data includes computing a plurality of intermediate vectors based on element-wise vector multiplication between different subsets of the vector element values and different respective pre-processing vectors. The vector-matrix multiplication module is loaded with a core matrix, and the input vector is multiplied by the model matrix based on separately multiplying each of the intermediate vectors by the loaded core matrix.
    Type: Application
    Filed: November 11, 2022
    Publication date: March 9, 2023
    Inventors: Matthew Raja Khoury, Rumen Rumenov Dangovski, Longwu Ou, Yichen Shen, Li Jing
  • Processing matrix operations for rate limited systems
    Patent number: 11526737
    Abstract: Data to be processed includes vector element values of an input vector and matrix element values of a model matrix associated with a neural network model. A vector-matrix multiplication module receives a set of matrix element values for performing a vector-matrix multiplication operation. Processing the data includes computing a plurality of intermediate vectors based on element-wise vector multiplication between different subsets of the vector element values and different respective pre-processing vectors. The vector-matrix multiplication module is loaded with a core matrix, and the input vector is multiplied by the model matrix based on separately multiplying each of the intermediate vectors by the loaded core matrix.
    Type: Grant
    Filed: January 31, 2020
    Date of Patent: December 13, 2022
    Assignee: Lightelligence, Inc.
    Inventors: Matthew Raja Khoury, Rumen Rumenov Dangovski, Longwu Ou, Yichen Shen, Li Jing
  • Processing Matrix Operations for Rate Limited Systems
    Publication number: 20200250518
    Abstract: Data to be processed includes vector element values of an input vector and matrix element values of a model matrix associated with a neural network model. A vector-matrix multiplication module receives a set of matrix element values for performing a vector-matrix multiplication operation. Processing the data includes computing a plurality of intermediate vectors based on element-wise vector multiplication between different subsets of the vector element values and different respective pre-processing vectors. The vector-matrix multiplication module is loaded with a core matrix, and the input vector is multiplied by the model matrix based on separately multiplying each of the intermediate vectors by the loaded core matrix.
    Type: Application
    Filed: January 31, 2020
    Publication date: August 6, 2020
    Inventors: Matthew Raja Khoury, Rumen Rumenov Dangovski, Longwu Ou, Yichen Shen, Li Jing