知搜-全球专利检索

  1. Login
  2. Sign Up

Search Results

17 records (took 0.015 seconds)

    Unsupervised representation learning and active learning to improve data efficiency
    1.
    发明授权
    Unsupervised representation learning and active learning to improve data efficiency 有权

    公开(公告)号：US12165311B2

    公开(公告)日：2024-12-10

    申请号：US17390408

    申请日：2021-07-30

    Applicant: SAMSUNG SDS AMERICA, INC.

    Inventor： Heng Hao ,  Sima Didari ,  Jae Oh Woo ,  Hankyu Moon ,  Patrick David Bangert

    IPC: G06F18/2411 ,  G06F16/28 ,  G06F16/55 ,  G06F18/21 ,  G06F18/214 ,  G06N3/08 ,  G06T7/00

    Abstract: A problem of imbalanced big data is solved by decoupling a classifier into a neural network for generation of representation vectors and into a classification model for operating on the representation vectors. The neural network and the classification model act as a mapper classifier. The neural network is trained with an unsupervised algorithm and the classification model is trained with a supervised active learning loop. An acquisition function is used in the supervised active learning loop to speed arrival at an accurate classification performance, improving data efficiency. The accuracy of the hybrid classifier is similar to or exceeds the accuracy of comparative classifiers in all aspects. In some embodiments, big data includes an imbalance of more than 10:1 in image classes. The hybrid classifier reduces labor and improves efficiency needed to arrive at an accurate classification performance, and improves recognition of previously-unrecognized images.

    OBJECT DISCOVERY
    2.
    发明申请
    OBJECT DISCOVERY 有权

    公开(公告)号：US20220383105A1

    公开(公告)日：2022-12-01

    申请号：US17517313

    申请日：2021-11-02

    Applicant: Samsung SDS America, Inc.

    Inventor： Hankyu MOON ,  Heng HAO ,  Sima DIDARI ,  Jae Oh WOO ,  Patrick David BANGERT

    IPC: G06N3/08 ,  G06K9/62

    Abstract: A problem of supervised learning is overcome by using patches to discover objects in unlabeled training images. The discovered objects are embedded in a pattern space. An AI machine replaces manual entry steps of training with a machine-centric process including clustering in a pixel space, clustering in latent space and building the pattern space based on different losses derived from pixel space clustering and latent space clustering. A distance structure in the pattern space captures the co-occurrence of patterns due to frequently appearing objects in training image data. Embodiments provide image representation based on local image patch naturally handles the position and scale invariance property that is important to effective object detection. Embodiments successfully identifies frequent objects such as human faces, human bodies, animals, or vehicles from unorganized data images based on a small quantity of training images.

    Device, method, and computer readable medium of generating recommendations via ensemble multi-arm bandit with an LPBoost
    3.
    发明授权
    Device, method, and computer readable medium of generating recommendations via ensemble multi-arm bandit with an LPBoost 有权

    公开(公告)号：US10796335B2

    公开(公告)日：2020-10-06

    申请号：US15290605

    申请日：2016-10-11

    Applicant: SAMSUNG SDS AMERICA, INC.

    Inventor： Qing He ,  Abhishek Mishra ,  Zhisu Zhu

    IPC: G06Q30/02

    Abstract: A method, an apparatus, and a computer readable medium of recommending contents. The method includes receiving, by a computer, at least one of user input and contextual input, wherein the contextual input corresponds to a plurality of arms, calculating, by the computer, a plurality of reward values for each of the plurality of arms using a plurality of individual recommendation algorithms such that each of the plurality of reward values is generated by a respective individual recommendation algorithm from the plurality of individual recommendation algorithms, based on the received input, calculating, by the computer, an aggregated reward value for each of the plurality of arms by applying linear program boosting to the plurality reward values for the respective arm; and selecting one arm from the plurality of arms which has greatest calculated aggregated reward value; and outputting, by the computer, contents corresponding to the selected arm.

    DEVICE, METHOD, AND COMPUTER READABLE MEDIUM OF GENERATING RECOMMENDATIONS VIA ENSEMBLE MULTI-ARM BANDIT WITH AN LPBOOST
    4.
    发明申请
    DEVICE, METHOD, AND COMPUTER READABLE MEDIUM OF GENERATING RECOMMENDATIONS VIA ENSEMBLE MULTI-ARM BANDIT WITH AN LPBOOST 审中-公开

    公开(公告)号：US20170103413A1

    公开(公告)日：2017-04-13

    申请号：US15290605

    申请日：2016-10-11

    Applicant: SAMSUNG SDS AMERICA, INC.

    Inventor： Qing HE ,  Abhishek MISHRA ,  Zhisu ZHU

    IPC: G06Q30/02

    CPC classification number: G06Q30/0244

    Abstract: A method, an apparatus, and a computer readable medium of recommending contents. The method includes receiving, by a computer, at least one of user input and contextual input, wherein the contextual input corresponds to a plurality of arms, calculating, by the computer, a plurality of reward values for each of the plurality of arms using a plurality of individual recommendation algorithms such that each of the plurality of reward values is generated by a respective individual recommendation algorithm from the plurality of individual recommendation algorithms, based on the received input, calculating, by the computer, an aggregated reward value for each of the plurality of arms by applying linear program boosting to the plurality reward values for the respective arm; and selecting one arm from the plurality of arms which has greatest calculated aggregated reward value; and outputting, by the computer, contents corresponding to the selected arm.

    SYSTEM AND METHOD FOR IDENTITY-BASED KEY MANAGEMENT
    6.
    发明申请
    SYSTEM AND METHOD FOR IDENTITY-BASED KEY MANAGEMENT 有权
    基于身份的关键管理系统与方法

    公开(公告)号：US20150039883A1

    公开(公告)日：2015-02-05

    申请号：US14447985

    申请日：2014-07-31

    Applicant: SAMSUNG SDS CO., LTD. ,  SDS AMERICA, INC.

    Inventor： Hyo-Jin YOON ,  Madjid NAKHJIRI

    IPC: H04L9/08 ,  H04L29/06 ,  H04L9/32

    CPC classification number: H04L9/0847 ,  H04L9/0863 ,  H04L9/3226 ,  H04L63/0442 ,  H04L63/061 ,  H04L63/083 ,  H04L63/0853

    Abstract: A system and method for identity (ID)-based key management are provided. The ID-based key management system includes an authentication server configured to authenticate a terminal through key exchange based on an ID and a password of a user of the terminal, set up a secure channel with the terminal, and provide a private key based on the ID of the user to the terminal through the secure channel, and a private-key generator configured to generate the private key corresponding to the ID of the terminal user according to a request of the authentication server.

    Abstract translation: 提供了基于身份（ID）的密钥管理的系统和方法。 基于ID的密钥管理系统包括认证服务器，其被配置为基于终端的用户的ID和密码通过密钥交换来认证终端，与终端建立安全信道，并且基于所述终端提供私钥 用户通过安全通道向终端的ID;以及私钥生成器，其被配置为根据认证服务器的请求生成与终端用户的ID对应的私钥。

    Continuously provisioning large-scale machine learning models
    8.
    发明授权
    Continuously provisioning large-scale machine learning models 有权

    公开(公告)号：US11514304B2

    公开(公告)日：2022-11-29

    申请号：US15606203

    申请日：2017-05-26

    Applicant: SAMSUNG SDS AMERICA, INC.

    Inventor： Jian Wu

    IPC: G06N3/08 ,  G06N20/00 ,  G06F8/71 ,  G06F30/20 ,  G06N5/00

    Abstract: An approach for continuously provisioning machine learning models, executed by one or more computer nodes to provide a future prediction in response to a request from one or more client devices, is provided. The approach generates, by the one or more computer nodes, a machine learning model. The approach determines, by the one or more computer nodes, whether the machine learning model is a new model. In response to determining the machine learning model is not the new model, the approach retrieves, by the one or more computer nodes, one or more model containers with an associated model to a new persistent model. The approach determines, by the one or more computer nodes, a difference between the associated model and the new persistent model. Further, in response to determining the machine learning model is the new model, the approach generates, by the one or more computer nodes, one or more model containers.

    CONTINUOUSLY PROVISIONING LARGE-SCALE MACHINE LEARNING MODELS
    9.
    发明申请
    CONTINUOUSLY PROVISIONING LARGE-SCALE MACHINE LEARNING MODELS 审中-公开

    公开(公告)号：US20170344910A1

    公开(公告)日：2017-11-30

    申请号：US15606203

    申请日：2017-05-26

    Applicant: SAMSUNG SDS AMERICA, INC.

    Inventor： Jian WU

    IPC: G06N99/00 ,  G06F17/30 ,  G06F17/50

    CPC classification number: G06N20/00 ,  G06F8/71 ,  G06F17/5009 ,  G06N5/003

    Abstract: An approach for continuously provisioning machine learning models, executed by one or more computer nodes to provide a future prediction in response to a request from one or more client devices, is provided. The approach generates, by the one or more computer nodes, a machine learning model. The approach determines, by the one or more computer nodes, whether the machine learning model is a new model. In response to determining the machine learning model is not the new model, the approach retrieves, by the one or more computer nodes, one or more model containers with an associated model to a new persistent model. The approach determines, by the one or more computer nodes, a difference between the associated model and the new persistent model. Further, in response to determining the machine learning model is the new model, the approach generates, by the one or more computer nodes, one or more model containers.

    CONTINUAL LEARNING IN SLOWLY-VARYING ENVIRONMENTS
    10.
    发明申请
    CONTINUAL LEARNING IN SLOWLY-VARYING ENVIRONMENTS 审中-公开

    公开(公告)号：US20170103341A1

    公开(公告)日：2017-04-13

    申请号：US15290409

    申请日：2016-10-11

    Applicant: SAMSUNG SDS AMERICA, INC.

    Inventor： KANNAN PARTHASARATHY ,  GIRISH KATHALAGIRI SOMASHEKARIAH ,  JOHN FRANCIS ARACKAPARAMBIL

    IPC: G06N99/00

    Abstract: An approach for continual learning in slowly-varying environments is provided. The approach receives one or more action requests from a decision agent. The approach deploys a first model to a decision engine. The approach initiates an observation period. The approach builds a second model, in which the second model comprises collected transaction data from the observation period. The approach initiates a test period. The approach determines a performance score for the first model and a performance score for the second model. The approach selects the model providing an optimized action.

Patent Agency Ranking