中科微点-全球专利检索

  1. Login
  2. Sign Up

Search Results

33 records (took 0.017 seconds)

    DATA PROCESSING SYSTEMS
    23.
    发明申请
    DATA PROCESSING SYSTEMS 审中-公开

    公开(公告)号：US20180232148A1

    公开(公告)日：2018-08-16

    申请号：US15432769

    申请日：2017-02-14

    Applicant: ARM Limited

    Inventor： Sharjeel Saeed

    IPC: G06F3/06

    CPC classification number: G06F12/00

    Abstract: Operating a data processing system including producing data in the form of plural blocks of data, where each block of data represents a particular region of an output data array, storing the data in a memory of the data processing system, and reading the data from the memory in the form of lines. Storing the data in the memory comprises storing each block of data of a first row of blocks of data in the memory at one or more memory addresses of a first set of memory addresses of a sequence of memory addresses for the memory, and storing each block of data of a second row of blocks of data in the memory at one or more memory addresses of a second set of different memory addresses of the sequence of memory addresses for the memory.

    DATA PROCESSING SYSTEMS
    24.
    发明申请
    DATA PROCESSING SYSTEMS 审中-公开

    公开(公告)号：US20170256027A1

    公开(公告)日：2017-09-07

    申请号：US15438663

    申请日：2017-02-21

    Applicant: ARM Limited

    Inventor： Sharjeel Saeed ,  Daren Croxford ,  Kushan Vijaykumar Vyas

    IPC: G06T1/60 ,  G06T1/20 ,  G06T3/60

    CPC classification number: G06T1/60 ,  G06T1/20 ,  G06T3/60 ,  G09G3/3611 ,  G09G5/00 ,  G09G5/363

    Abstract: A method of operating a data processing system 4 comprises a first processing stage 11, 12 of the data processing system producing data according to a first pattern, and a second processing stage 20 of the data processing system using the data produced by the first processing stage 11, 12 according to a second different pattern. The data processing system 4 deactivates the first processing stage 11, 12 when the first processing stage 11, 12 has produced a set of data that includes sufficient data to allow the set of data to be used by the second processing stage 20 according to the second pattern, and re-activates the first processing stage 11, 12 based on the use of that set of data by the second processing stage 20.

    Convolution size prediction to reduce calculations
    25.
    发明授权
    Convolution size prediction to reduce calculations 有权

    公开(公告)号：US12299567B2

    公开(公告)日：2025-05-13

    申请号：US17479257

    申请日：2021-09-20

    Applicant: Arm Limited

    Inventor： Daren Croxford ,  Sharjeel Saeed ,  Rachel Jean Trimble

    IPC: G06N3/04 ,  G06F18/10 ,  G06F18/211 ,  G06N3/0464 ,  G06N3/08 ,  G06V10/75

    Abstract: There is provided a data processing apparatus for performing machine learning. The data processing apparatus includes convolution circuitry for convolving a plurality of neighbouring regions of input data using a kernel to produce convolution outputs. Max-pooling circuitry determines and selects the largest of the convolution outputs as a pooled output and prediction circuitry performs a size prediction of the convolution outputs based on the neighbouring regions, wherein the size prediction is performed prior to the max-pooling circuitry determining the largest of the convolution outputs and adjusts a behaviour of the convolution circuitry based on the size prediction.

    NEURAL NETWORK PROCESSING
    27.
    发明公开
    NEURAL NETWORK PROCESSING 审中-公开

    公开(公告)号：US20230316063A1

    公开(公告)日：2023-10-05

    申请号：US17708474

    申请日：2022-03-30

    Applicant: Arm Limited

    Inventor： Rachel Jean Trimble ,  Sharjeel Saeed ,  Daren Croxford

    IPC: G06N3/08 ,  G06N3/04 ,  G06K9/62

    CPC classification number: G06N3/08 ,  G06N3/04 ,  G06K9/6201

    Abstract: An input data array is subjected to neural network processing to generate a result of the neural network processing for the input data array. A perturbation is applied to a part (but not all of) the input data array, with neural network processing then performed using the so-perturbed version of the input data array. However only some (and not all) of the perturbed version is subjected to neural network processing, based on the part of the input data array to which the perturbation has been applied. The result of the neural network processing of the perturbed version of the input data array is compared with the result of the neural network processing of the input data array without the perturbation, to determine whether the perturbation of the input data array has an effect on the result of the neural network processing.

    Neural network processing
    28.
    发明授权
    Neural network processing 有权

    公开(公告)号：US11625578B2

    公开(公告)日：2023-04-11

    申请号：US16834881

    申请日：2020-03-30

    Applicant: Arm Limited ,  Apical Limited

    Inventor： Sharjeel Saeed ,  Aaron DeBattista ,  Daren Croxford

    IPC: G06N3/04 ,  G06K9/62 ,  G06N3/08

    Abstract: A method apparatus and computer readable medium for processing input data using a neural network comprising at least a first layer and a second layer. The method comprising the steps of applying a partitioning scheme to the input data, to partition the input data into a plurality of blocks, each block representing a portion of the input data. At the first layer of the neural network, the blocks of the input data are processed in a first order to generate intermediary data, wherein the intermediary data is partitioned into a plurality of intermediary blocks. At the second layer of the neural network, the intermediary blocks are processed in a second order, wherein the second order differs from the first order.

    Data processing method and system for performing convolutions
    29.
    发明授权
    Data processing method and system for performing convolutions 有权

    公开(公告)号：US11423117B2

    公开(公告)日：2022-08-23

    申请号：US16552548

    申请日：2019-08-27

    Applicant: Arm Limited ,  Apical Limited

    Inventor： Sharjeel Saeed ,  Daren Croxford ,  Davide Marani ,  Jayavarapu Srinivasa Rao

    IPC: G06F17/15 ,  G06F1/32 ,  G06N3/04 ,  G06F7/544

    Abstract: A computer implemented method for performing convolutions between subsets of an input data array and a kernel resulting in subsets of an output data array. The method may include receiving an input data array and using positional data indicating the position of elements of the input data array to determine subsets of the input data array which contains at least one non-zero value data element; performing convolutions between the subsets of the input data array containing at least one non-zero value data element and a kernel to produce output data array subsets; and combining the output data subsets with the positional data to generate output data indicative of a completed output data array.

    Processing input data in a convolutional neural network
    30.
    发明授权
    Processing input data in a convolutional neural network 有权

    公开(公告)号：US10853694B2

    公开(公告)日：2020-12-01

    申请号：US16209505

    申请日：2018-12-04

    Applicant: Apical Ltd ,  Arm Limited

    Inventor： Daren Croxford ,  Sharjeel Saeed ,  Sean Tristram LeGuay Ellis

    IPC: G06K9/36 ,  G06K9/62 ,  G06N3/08 ,  G06N3/04

    Abstract: A method of processing input data using a computing system. The method comprises obtaining association data which relates a kernel in a convolutional neural network to one or more known data patterns; conducting analysis of input data for the convolutional neural network to identify whether a region of input data corresponds to at least one of the one or more known data patterns; and determining whether to process the region of input data with the kernel in the convolutional neural network based on the analysis and the association data.

Patent Agency Ranking