中科微点-全球专利检索

  1. Login
  2. Sign Up

Search Results

55 records (took 0.024 seconds)

    Cross-model score normalization
    3.
    发明授权
    Cross-model score normalization 有权

    公开(公告)号：US11995521B2

    公开(公告)日：2024-05-28

    申请号：US16681599

    申请日：2019-11-12

    Applicant: Dropbox, Inc.

    Inventor： Jongmin Baek

    IPC: G06N20/00 ,  G06F16/16 ,  G06F16/93 ,  G06N5/04 ,  H04L67/10

    CPC classification number: G06N20/00 ,  G06F16/16 ,  G06F16/93 ,  G06N5/04 ,  H04L67/10

    Abstract: Computer-implemented techniques encompass using distinct machine learning sub-models to score respective types of candidate content for the purpose of providing personalized content suggestions to end-users of a content management system. The relevancy scores generated by the distinct sub-models are mapped to expected end-user interaction scores of the candidate content scored. Content suggestions are provided at end-users' computing devices where the suggested content is selected from the candidate content based on the expected end-user interaction scores of the candidate content. For each distinct sub-model, a normalizing mapping function is solved using an optimizer that maps the relevancy scores generated by the sub-model for the candidate content to expected end-user interaction scores for the candidate content. The expected end-user interaction scores are comparable across the distinct sub-models and can be used to rank content suggestions across the distinct sub-models.

    UTILIZING MACHINE-LEARNING MODELS TO GENERATE IDENTIFIER EMBEDDINGS AND DETERMINE DIGITAL CONNECTIONS BETWEEN DIGITAL CONTENT ITEMS
    4.
    发明公开
    UTILIZING MACHINE-LEARNING MODELS TO GENERATE IDENTIFIER EMBEDDINGS AND DETERMINE DIGITAL CONNECTIONS BETWEEN DIGITAL CONTENT ITEMS 审中-公开

    公开(公告)号：US20230169139A1

    公开(公告)日：2023-06-01

    申请号：US18153960

    申请日：2023-01-12

    Applicant: Dropbox, Inc.

    Inventor： Jongmin Baek ,  Jiarui Ding ,  Ermo Wei ,  Scott McCrae

    IPC: G06F16/958 ,  G06N20/00 ,  G06F40/284 ,  G06N3/045

    CPC classification number: G06F16/958 ,  G06F40/284 ,  G06N3/045 ,  G06N20/00

    Abstract: The present disclosure relates to systems, methods, and non-transitory computer-readable media that utilize machine learning models to generate identifier embeddings from digital content identifiers and then leverage these identifier embeddings to determine digital connections between digital content items. In particular, the disclosed systems can utilize an embedding machine-learning model that comprises a character-level embedding machine-learning model and a word-level embedding machine-learning model. For example, the disclosed systems can combine a character embedding from the character-level embedding machine-learning model and a token embedding from the word-level embedding machine-learning model. The disclosed systems can determine digital connections between the plurality of digital content items by processing these identifier embeddings for a plurality of digital content items utilizing a content management model. Based on the digital connections, the disclosed systems can surface one or more digital content suggestions to a user interface of a client device.

    Utilizing a natural language model to determine a predicted activity event based on a series of sequential tokens
    5.
    发明授权
    Utilizing a natural language model to determine a predicted activity event based on a series of sequential tokens 有权

    公开(公告)号：US11567812B2

    公开(公告)日：2023-01-31

    申请号：US17065266

    申请日：2020-10-07

    Applicant: Dropbox, Inc.

    Inventor： Ranjitha Gurunath Kulkarni ,  Xingyu Xiang ,  Jongmin Baek ,  Ermo Wei

    IPC: G06F9/54 ,  G06F40/284 ,  G06N3/08 ,  G06N5/02

    Abstract: The present disclosure relates to systems, methods, and non-transitory computer-readable media that can leverage a natural language model to determine a most probable candidate sequence of tokens and thereby generate a predicted user activity. In particular, the disclosed systems can tokenize activity event vectors to generate a series of sequential tokens that correspond to recent user activity of one or more user accounts. In addition, the disclosed systems can, for each candidate (e.g., hypothetical) user activity, augment the series of sequential tokens to include a corresponding token. Based on respective probability scores for each of the augmented series of sequential tokens, the disclosed systems can identify as the predicted user activity, a candidate user activity corresponding to one of the augmented series of sequential tokens associated with a highest probability score. Based on the predicted user activity, the disclosed systems can surface one or more suggestions to a client device.

    Data storage scheme switching in a distributed data storage system
    6.
    发明授权
    Data storage scheme switching in a distributed data storage system 有权

    公开(公告)号：US11422721B2

    公开(公告)日：2022-08-23

    申请号：US16839605

    申请日：2020-04-03

    Applicant: Dropbox, Inc.

    Inventor： Michael Loh ,  Daniel R. Horn ,  Andraz Kavalar ,  David Lichtenberg ,  Austin Sung ,  Shi Feng ,  Jongmin Baek

    IPC: G06F3/06 ,  G06N20/00 ,  G06N7/00 ,  H03M13/15

    Abstract: Systems and methods for dynamic and automatic data storage scheme switching in a distributed data storage system. A machine learning-based policy for computing probable future content item access patterns based on historical content item access patterns is employed to dynamically and automatically switch the storage of content items (e.g., files, digital data, photos, text, audio, video, streaming content, cloud documents, etc.) between different data storage schemes. The different data storage schemes may have different data storage cost and different data access cost characteristics. For example, the different data storage schemes may encompass different types of data storage devices, different data compression schemes, and/or different data redundancy schemes.

    Enhancing a digital image
    7.
    发明授权
    Enhancing a digital image 有权

    公开(公告)号：US11334970B2

    公开(公告)日：2022-05-17

    申请号：US16866324

    申请日：2020-05-04

    Applicant: Dropbox, Inc.

    Inventor： Jongmin Baek

    IPC: G06T5/00 ,  G06T7/11 ,  G06T7/168 ,  G06T5/20 ,  G06T7/12 ,  G06T7/136 ,  G06T1/00

    Abstract: One or more embodiments of an image enhancement system enable a computing device to generate an enhanced digital image. In particular, a computing device can enhance a digital image including, for example, a photograph of a whiteboard, document, chalkboard, or other object having a uniform background. The computing device can determine modifications to apply to the digital image by minimizing an energy heuristic that both causes pixels of the digital image to change to a uniform color (e.g., white) and preserves gradients from the digital image. The computing device can further generate an enhanced digital image by applying the determined modifications to the digital image.

    Lossless compression of a content item using a neural network trained on content item cohorts
    10.
    发明授权
    Lossless compression of a content item using a neural network trained on content item cohorts 有权

    公开(公告)号：US10177783B1

    公开(公告)日：2019-01-08

    申请号：US15947768

    申请日：2018-04-06

    Applicant: Dropbox, Inc.

    Inventor： Daniel R. Horn ,  Jongmin Baek ,  Anatoly Yakovenko

    IPC: H03M7/00 ,  H03M7/30 ,  H03M7/40 ,  G06N3/04 ,  G06N3/08

    Abstract: Lossless compression of a content item using a neural network trained on content item cohorts. A computing system includes a neural network that is used to train a plurality of symbol prediction models. Each symbol prediction model is trained based on a corresponding cohort of content items. A particular symbol prediction model of the models trained is selected based on an intrinsic characteristic of a particular content item to be losslessly compressed such as, for example, the type or file extension of the content item. The content item is then losslessly compressed based on a set of symbol predictions fed to an arithmetic coder that are generated using the particular symbol prediction model selected.

Patent Agency Ranking