中科微点-全球专利检索

  1. Login
  2. Sign Up

Search Results

12 records (took 0.016 seconds)

    USING MULTIPLE TRAINED MODELS TO REDUCE DATA LABELING EFFORTS
    4.
    发明公开
    USING MULTIPLE TRAINED MODELS TO REDUCE DATA LABELING EFFORTS 审中-公开

    公开(公告)号：US20240281431A1

    公开(公告)日：2024-08-22

    申请号：US18647425

    申请日：2024-04-26

    Applicant: Xerox Corporation

    Inventor： Matthew Shreve ,  Francisco E. Torres ,  Raja Bala ,  Robert R. Price ,  Pei Li

    IPC: G06F16/23 ,  G06N20/00

    CPC classification number: G06F16/2379 ,  G06N20/00

    Abstract: A method of labeling training data includes inputting a plurality of unlabeled input data samples into each of a plurality of pre-trained neural networks and extracting a set of feature embeddings from multiple layer depths of each of the plurality of pre-trained neural networks. The method also includes generating a plurality of clusterings from the set of feature embeddings. The method also includes analyzing, by a processing device, the plurality of clusterings to identify a subset of the plurality of unlabeled input data samples that belong to a same unknown class. The method also includes assigning pseudo-labels to the subset of the plurality of unlabeled input data samples.

    Using multiple trained models to reduce data labeling efforts
    7.
    发明授权
    Using multiple trained models to reduce data labeling efforts 有权

    公开(公告)号：US11983171B2

    公开(公告)日：2024-05-14

    申请号：US18219333

    申请日：2023-07-07

    Applicant: Xerox Corporation

    Inventor： Matthew Shreve ,  Francisco E. Torres ,  Raja Bala ,  Robert R. Price ,  Pei Li

    IPC: G06F16/00 ,  G06F16/23 ,  G06N20/00

    CPC classification number: G06F16/2379 ,  G06N20/00

    Abstract: A method of labeling a dataset includes inputting a testing set comprising a plurality of input data samples into a plurality of pre-trained machine learning models to generate a set of embeddings output by the plurality of pre-trained machine learning models. The method further includes performing an iterative cluster labeling algorithm that includes generating a plurality of clusterings from the set of embeddings, analyzing the plurality of clusterings to identify a target embedding with a highest duster quality, analyzing the target embedding to determine a compactness for each of the plurality of clusterings of the target embedding, and identifying a target cluster among the plurality of clusterings of the target embedding based on the compactness. The method further includes assigning pseudo-labels to the subset of the plurality of input data samples that are members of the target duster.

    USING MULTIPLE TRAINED MODELS TO REDUCE DATA LABELING EFFORTS
    10.
    发明公开
    USING MULTIPLE TRAINED MODELS TO REDUCE DATA LABELING EFFORTS 审中-公开

    公开(公告)号：US20230350880A1

    公开(公告)日：2023-11-02

    申请号：US18219333

    申请日：2023-07-07

    Applicant: Xerox Corporation

    Inventor： Matthew Shreve ,  Francisco E. Torres ,  Raja Bala ,  Robert R. Price ,  Pei Li

    IPC: G06F16/23 ,  G06N20/00

    CPC classification number: G06F16/2379 ,  G06N20/00

    Abstract: A method of labeling a dataset includes inputting a testing set comprising a plurality of input data samples into a plurality of pre-trained machine learning models to generate a set of embeddings output by the plurality of pre-trained machine learning models. The method further includes performing an iterative cluster labeling algorithm that includes generating a plurality of clusterings from the set of embeddings, analyzing the plurality of clusterings to identify a target embedding with a highest duster quality, analyzing the target embedding to determine a compactness for each of the plurality of clusterings of the target embedding, and identifying a target cluster among the plurality of clusterings of the target embedding based on the compactness. The method further includes assigning pseudo-labels to the subset of the plurality of input data samples that are members of the target duster.

Patent Agency Ranking