知搜-全球专利检索

  1. Login
  2. Sign Up

Search Results

74 records (took 0.035 seconds)

    DYNAMIC LOW-RANK ESTIMATION FOR TRANSFORMER-BASED LANGUAGE MODELS
    1.
    发明申请
    DYNAMIC LOW-RANK ESTIMATION FOR TRANSFORMER-BASED LANGUAGE MODELS 有权

    公开(公告)号：US20250029005A1

    公开(公告)日：2025-01-23

    申请号：US18669413

    申请日：2024-05-20

    Applicant: Samsung Electronics Co., Ltd.

    Inventor： Ting Hua ,  Xiao Li ,  Shangqian Gao ,  Yen-Chang Hsu ,  Yilin Shen ,  Hongxia Jin

    IPC: G06N20/00 ,  G10L15/18 ,  G10L15/26

    Abstract: A method includes accessing a plurality of weight matrices of a machine learning model. The method also includes, for each weight matrix, decomposing the weight matrix into a U matrix, an S matrix, and a V matrix using singular value decomposition. The S matrix is a diagonal matrix, and a singular group corresponds to each element in the S matrix. The method further includes, for each weight matrix, determining an importance score of each singular group. The importance score of the singular group represents a change in loss if the singular group is removed from the machine learning model. The method also includes, for each weight matrix, ranking the singular groups across the plurality of weight matrices based on the importance scores. In addition, the method includes, for each weight matrix, identifying one or more of the singular groups to prune based on the ranking of the singular groups.

    SYSTEM AND METHOD FOR CONTINUAL REFINABLE NETWORK
    2.
    发明公开
    SYSTEM AND METHOD FOR CONTINUAL REFINABLE NETWORK 审中-公开

    公开(公告)号：US20230177332A1

    公开(公告)日：2023-06-08

    申请号：US18061216

    申请日：2022-12-02

    Applicant: Samsung Electronics Co., Ltd.

    Inventor： Sima Behpour ,  Yilin Shen ,  Hongxia Jin

    IPC: G06N3/08

    CPC classification number: G06N3/08

    Abstract: A method includes accessing, using at least one processor of an electronic device, a machine learning model. The machine learning model is trained by directing a gradient direction of gradients to one or more flat local minima and using a dynamic learning rate for one or more additional tasks. The method also includes receiving, using the at least one processor, an input from an input source. The method further includes providing, using the at least one processor, the input to the machine learning model. The method also includes receiving, using the at least one processor, an output from the machine learning model. In addition, the method includes instructing, using the at least one processor, at least one action based on the output from the machine learning model.

    Generating annotated natural language phrases
    4.
    发明授权
    Generating annotated natural language phrases 有权

    公开(公告)号：US11036926B2

    公开(公告)日：2021-06-15

    申请号：US16236886

    申请日：2018-12-31

    Applicant: Samsung Electronics Co., Ltd.

    Inventor： Yilin Shen ,  Avik Ray ,  Hongxia Jin

    IPC: G06F40/169 ,  G06F40/40 ,  G06F40/30 ,  G06F40/20 ,  G06F16/332

    Abstract: A system receives a phrase that includes at least one tagged object and generates instantiated phrases by instantiations of each tagged object in the phrase. The system generates lists of natural language phrases by corresponding paraphrases of each of the instantiated phrases. The system generates ordered lists of natural language phrases by ordering natural language phrases in each list of natural language phrases based on occurrences of each natural language phrase. The system generates annotated natural language phrases by using each tagged object in the phrase to annotate the ordered lists of natural language phrases or an enhanced set of natural language phrases that is based on the ordered lists of natural language phrases.

    ON-DEVICE LIGHTWEIGHT NATURAL LANGUAGE UNDERSTANDING (NLU) CONTINUAL LEARNING
    5.
    发明申请
    ON-DEVICE LIGHTWEIGHT NATURAL LANGUAGE UNDERSTANDING (NLU) CONTINUAL LEARNING 有权

    公开(公告)号：US20210004532A1

    公开(公告)日：2021-01-07

    申请号：US16946746

    申请日：2020-07-02

    Applicant: Samsung Electronics Co., Ltd.

    Inventor： Yilin Shen ,  Xiangyu Zeng ,  Hongxia Jin

    IPC: G06F40/279

    Abstract: A method includes obtaining, using at least one processor of an electronic device, a base model trained to perform natural language understanding. The method also includes generating, using the at least one processor, a first model expansion based on knowledge from the base model. The method further includes training, using the at least one processor, the first model expansion based on first utterances without modifying parameters of the base model. The method also includes receiving, using the at least one processor, an additional utterance from a user. In addition, the method includes determining, using the at least one processor, a meaning of the additional utterance using the base model and the first model expansion.

    MULTI-MODELS THAT UNDERSTAND NATURAL LANGUAGE PHRASES
    6.
    发明申请
    MULTI-MODELS THAT UNDERSTAND NATURAL LANGUAGE PHRASES 审中-公开

    公开(公告)号：US20190332668A1

    公开(公告)日：2019-10-31

    申请号：US16390241

    申请日：2019-04-22

    Applicant: Samsung Electronics Co., Ltd.

    Inventor： Yu Wang ,  Yilin Shen ,  Hongxia Jin

    IPC: G06F17/27

    Abstract: A system determines intent values based on an object in a received phrase, and detail values based on the object in the received phrase. The system determines intent state values based on the intent values and the detail values, and detail state values and an intent detail value based on the intent values and the detail values. The system determines other intent values based on the intent values and another object in the received phrase, and other detail values based on the detail values and the other object in the received phrase. The system determines a general intent value based on the other intent values, the other detail values, and the intent state values, and another intent detail value based on the other intent values, the other detail values, and the detail state values.

    SYSTEM AND METHOD FOR ACTIVE MACHINE LEARNING
    7.
    发明申请
    SYSTEM AND METHOD FOR ACTIVE MACHINE LEARNING 审中-公开

    公开(公告)号：US20190318261A1

    公开(公告)日：2019-10-17

    申请号：US16370542

    申请日：2019-03-29

    Applicant: Samsung Electronics Co., Ltd.

    Inventor： Yue Deng ,  Yilin Shen ,  Hongxia Jin

    IPC: G06N20/00 ,  G06K9/62

    Abstract: An electronic device for active learning includes at least one memory and at least one processor coupled to the at least one memory. The at least one processor is configured to select one or more entries from a data set including unlabeled data based on a similarity between the one or more entries and labeled data. The at least one processor is further configured to cause the one or more entries to be labeled.

    METHOD TO LEARN PERSONALIZED INTENTS
    8.
    发明申请
    METHOD TO LEARN PERSONALIZED INTENTS 审中-公开

    公开(公告)号：US20190266237A1

    公开(公告)日：2019-08-29

    申请号：US15904203

    申请日：2018-02-23

    Applicant: Samsung Electronics Co., Ltd.

    Inventor： Avik Ray ,  Yilin Shen ,  Hongxia Jin

    IPC: G06F17/27 ,  G06F3/16 ,  G10L15/07 ,  G10L15/18

    Abstract: A method includes retrieving, at an electronic device, a first natural language (NL) input. An intent of the first NL input is undetermined by both a generic parser and a personal parser. A paraphrase of the first NL input is retrieved at the electronic device. An intent of the paraphrase of the first NL input is determined using at least one of: the generic parser, the personal parser, or a combination thereof. A new personal intent for the first NL input is generated based on the determined intent. The personal parser is trained using existing personal intents and the new personal intent.

    SYSTEM AND METHOD TO ENABLE PRIVACY-PRESERVING REAL TIME SERVICES AGAINST INFERENCE ATTACKS
    9.
    发明申请
    SYSTEM AND METHOD TO ENABLE PRIVACY-PRESERVING REAL TIME SERVICES AGAINST INFERENCE ATTACKS 审中-公开

    公开(公告)号：US20170220817A1

    公开(公告)日：2017-08-03

    申请号：US15011368

    申请日：2016-01-29

    Applicant: Samsung Electronics Co., Ltd.

    Inventor： Yilin Shen ,  Hongxia Jin

    IPC: G06F21/62 ,  H04L29/06

    Abstract: One embodiment provides a method comprising receiving general private data identifying at least one type of privacy-sensitive data to protect, collecting at least one type of real-time data, and determining an inference privacy risk level associated with transmitting the at least one type of real-time data to a second device. The inference privacy risk level indicates a degree of risk of inferring the general private data from transmitting the at least one type of real-time data. The method further comprises distorting at least a portion of the at least one type of real-time data based on the inference privacy risk level before transmitting the at least one type of real-time data to the second device.

Patent Agency Ranking