中科微点-全球专利检索

  1. Login
  2. Sign Up

Search Results

151 records (took 0.039 seconds)

    AUTOMATED DATA PRE-PROCESSING FOR MACHINE LEARNING
    1.
    发明公开
    AUTOMATED DATA PRE-PROCESSING FOR MACHINE LEARNING 审中-公开

    公开(公告)号：US20240265664A1

    公开(公告)日：2024-08-08

    申请号：US18166004

    申请日：2023-02-08

    Applicant: International Business Machines Corporation

    Inventor： Yu Zui You ,  Zhan Peng Huo ,  Jun Zhu ,  Kuo-Liang Chou ,  Xuan Feng ,  JUN HAO

    IPC: G06V10/20 ,  G06V10/774

    CPC classification number: G06V10/20 ,  G06V10/774

    Abstract: Computer-implemented methods for performing automated pre-processing of data for a machine-learning based prediction system are provided. Aspects include receiving a plurality of raw data sets, receiving a plurality of processed data sets, wherein each of the plurality of processed data sets corresponds to one of the plurality of raw data sets, and generating a plurality of pre-processing templates based on the plurality of raw data sets and the processed data set. Aspects also include receiving an input data set, generating, for each of the plurality of pre-processing templates, a matching score for the input data set, and selecting one of the plurality of pre-processing templates based on the matching score. Aspects further include pre-processing the input data set using the selected pre-processing template and providing the pre-processed input data set to the machine learning based prediction system.

    Spatial computing for location-based services
    2.
    发明授权
    Spatial computing for location-based services 有权

    公开(公告)号：US11748387B2

    公开(公告)日：2023-09-05

    申请号：US17133935

    申请日：2020-12-24

    Applicant: International Business Machines Corporation

    Inventor： Ning Duan ,  Yu Huang ,  Zhi Hu Wang ,  Shiping Yan ,  Xin Zhang ,  Jun Zhu

    IPC: G06F16/00 ,  G06F16/29 ,  G06F16/901

    CPC classification number: G06F16/29 ,  G06F16/9027

    Abstract: Techniques facilitating resolution-based spatial computing are provided. In one example, a computer-implemented method comprises traversing, by a device operatively coupled to a processor, a data structure corresponding to a land area for a location having an index; and determining, by the device, whether the location is at least partially within the land area based on a result of the traversing. In some embodiments, the traversing comprises: obtaining a threshold number of levels based at least in part on a resolution parameter; scanning a first level of the data structure for a node having an index corresponding to the index of the location; and iterating the scanning for respective subsequent levels of the data structure based on the scanning returning a node having subordinate nodes and a number of levels for which the scanning and iterating have been performed being less than the threshold number of levels.

    Method, system, and manufacture for light hypergraph based recommendation
    4.
    发明授权
    Method, system, and manufacture for light hypergraph based recommendation 有权

    公开(公告)号：US11334935B2

    公开(公告)日：2022-05-17

    申请号：US17012420

    申请日：2020-09-04

    Applicant: International Business Machines Corporation

    Inventor： Chao Chen ,  Dong Sheng Li ,  Stephen My Chu ,  Jun Zhu

    IPC: G06Q30/00 ,  G06Q30/06 ,  G06F16/901

    Abstract: Aspects of the invention include a computer-implemented method for generating product recommendations used a hypergraph. The computer-implemented method includes generating a product embedding vector for a product unpurchased by a consumer based on a hypergraph. Generating a consumer embedding vector for the consumer based on the hypergraph. An affinity between the consumer and a product unpurchased by the consumer is scored based on the product embedding vector and the consumer embedding vector. A digital description of the unpurchased product is retrieved based on the scoring. The digital description is presented to the consumer.

    Feature extraction using multi-task learning
    5.
    发明授权
    Feature extraction using multi-task learning 有权

    公开(公告)号：US11100399B2

    公开(公告)日：2021-08-24

    申请号：US15818877

    申请日：2017-11-21

    Applicant: INTERNATIONAL BUSINESS MACHINES CORPORATION

    Inventor： Wei Shan Dong ,  Peng Gao ,  Chang Sheng Li ,  Chun Yang Ma ,  Kai AD Yang ,  Ren Jie Yao ,  Ting Yuan ,  Jun Zhu

    IPC: G06N3/08 ,  G06N3/04

    Abstract: Systems and methods for training a neural network model are disclosed. In the method, training data is obtained by a deep neural network (DNN) first, the deep neural network comprising at least one hidden layer. Then features of the training data are obtained from a specified hidden layer of the at least one hidden layer, the specified hidden layer being connected respectively to a supervised classification network for classification tasks and an autoencoder based reconstruction network for reconstruction tasks. And at last the DNN, the supervised classification network and the reconstruction network are trained as a whole based on the obtained features, the training being guided by the classification tasks and the reconstruction tasks.

    CLASSIFYING IMAGES IN OVERLAPPING GROUPS OF IMAGES USING CONVOLUTIONAL NEURAL NETWORKS
    7.
    发明申请
    CLASSIFYING IMAGES IN OVERLAPPING GROUPS OF IMAGES USING CONVOLUTIONAL NEURAL NETWORKS 有权

    公开(公告)号：US20210158094A1

    公开(公告)日：2021-05-27

    申请号：US16692356

    申请日：2019-11-22

    Applicant: International Business Machines Corporation

    Inventor： Peng JI ,  Guo Qiang Hu ,  Yuan Yuan Ding ,  Jun Zhu ,  Jing Chang Huang ,  Sheng Nan Zhu

    IPC: G06K9/62 ,  G06Q50/04 ,  G06N3/08 ,  G06N3/04

    Abstract: The present disclosure relates to training a machine learning model to classify images. An example method generally includes receiving a training data set including images in a first category and images in a second category. A convolutional neural network (CNN) is trained using the training data set, and a feature map is generated from layers of the CNN based on features of images in the training data set. A first area in the feature map including images in the first category and a second area in the feature map where images in the first category overlap with images in the second category are identified. The first category is split into a first subcategory corresponding to the first area and a second subcategory corresponding to the second area. The CNN is retrained based on the images in the first subcategory, images in the second subcategory, and images in the second category.

    IMAGE RECOVERY
    8.
    发明申请
    IMAGE RECOVERY 有权

    公开(公告)号：US20210118113A1

    公开(公告)日：2021-04-22

    申请号：US16654962

    申请日：2019-10-16

    Applicant: International Business Machines Corporation

    Inventor： Fan Li ,  Guo Qiang Hu ,  Jun Zhu ,  Sheng Nan Zhu ,  JingChang Huang ,  Yuan Yuan Ding ,  Peng Ji

    IPC: G06T7/00 ,  G06K9/62 ,  G06T11/60 ,  G06F17/18

    Abstract: A method, a device and a computer program product for image processing are proposed. In the method, a first training image and region information are obtained. The region information indicates a region of a defect in the first training image. A second training image with the defect at least partially removed is generated using an image generator based on the first training image and the region information. The image generator is trained to recover the first training image by replacing pixels included in the region indicated by the region information. The image generator is updated based on the second training image. In this way, the image including the defect can be accurately and efficiently recovered.

    Self-guided object detection in regular images
    9.
    发明授权
    Self-guided object detection in regular images 有权

    公开(公告)号：US10956796B2

    公开(公告)日：2021-03-23

    申请号：US16157418

    申请日：2018-10-11

    Applicant: INTERNATIONAL BUSINESS MACHINES CORPORATION

    Inventor： Jun Chi Yan ,  Jun Zhu ,  Guo Qiang Hu ,  Jing Chang Huang ,  Chang Chun Liu ,  Zhi Hu Wang ,  Peng Ji

    IPC: G06K9/68 ,  G06K9/62 ,  G06K9/00

    Abstract: A computer-implemented method is provided for image-based, self-guided object detection. The method includes receiving, by a processor device, a set of images. Each of the images has a respective grid thereon that is labeled regarding a respective object to be detected using grid level label data. The method further includes training, by the processor device, a grid-based object detector using the grid level label data. The method also includes determining, by the processor device, a respective bounding box for the respective object in each of the images, by applying local segmentation to each of the images. The method additionally includes training, by the processor device, a Region-based Convolutional Neural Network (RCNN) for joint object localization and object classification using the respective bounding box for the respective object in each of the images as an input to the RCNN.

Patent Agency Ranking