Patents by Inventor Xi Ge

Xi Ge 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).

  • Patent number: 11568268
    Abstract: A deep learning heterogeneous computing method based on layer-wide memory allocation, at least comprises steps of: traversing a neural network model so as to acquire a training operational sequence and a number of layers L thereof; calculating a memory room R1 required by data involved in operation at the ith layer of the neural network model under a double-buffer configuration, where 1?i?L; altering a layer structure of the ith layer and updating the training operational sequence; distributing all the data across a memory room of the CPU and the memory room of the GPU according to a data placement method; performing iterative computation at each said layer successively based on the training operational sequence so as to complete neural network training.
    Type: Grant
    Filed: January 21, 2020
    Date of Patent: January 31, 2023
    Assignee: HUAZHONG UNIVERSITY OF SCIENCE AND TECHNOLOGY
    Inventors: Hai Jin, Xiaofei Liao, Long Zheng, Haikun Liu, Xi Ge
  • Publication number: 20200272907
    Abstract: A deep learning heterogeneous computing method based on layer-wide memory allocation, at least comprises steps of: traversing a neural network model so as to acquire a training operational sequence and a number of layers L thereof; calculating a memory room R1 required by data involved in operation at the ith layer of the neural network model under a double-buffer configuration, where 1?i?L; altering a layer structure of the ith layer and updating the training operational sequence; distributing all the data across a memory room of the CPU and the memory room of the GPU according to a data placement method; performing iterative computation at each said layer successively based on the training operational sequence so as to complete neural network training.
    Type: Application
    Filed: January 21, 2020
    Publication date: August 27, 2020
    Inventors: Hai JIN, Xiaofei LIAO, Long ZHENG, Haikun LIU, Xi GE
  • Patent number: 9377678
    Abstract: A method of processing a semiconductor wafer may include providing a rotatably alignable photolithography mask that includes different mask images. Each mask image may be in a corresponding different mask sector. The method may also include performing a series of exposures with the rotatably alignable photolithography mask at different rotational alignments with respect to the semiconductor wafer so that the different mask images produce at least one working semiconductor wafer sector, and at least one non-working semiconductor wafer sector.
    Type: Grant
    Filed: November 20, 2013
    Date of Patent: June 28, 2016
    Assignee: STMICROELECTRONICS PTE LTD
    Inventors: Alan Lee, Xi Ge
  • Publication number: 20150140479
    Abstract: A method of processing a semiconductor wafer may include providing a rotatably alignable photolithography mask that includes different mask images. Each mask image may be in a corresponding different mask sector. The method may also include performing a series of exposures with the rotatably alignable photolithography mask at different rotational alignments with respect to the semiconductor wafer so that the different mask images produce at least one working semiconductor wafer sector, and at least one non-working semiconductor wafer sector.
    Type: Application
    Filed: November 20, 2013
    Publication date: May 21, 2015
    Applicant: STMICROELECTRONICS PTE LTD
    Inventors: Alan Lee, Xi Ge