中科微点-全球专利检索

  1. Login
  2. Sign Up

Search Results

35254 records (took 0.046 seconds)

    MACHINE LEARNING FOR IDENTIFYING IDLE SESSIONS
    1.
    发明公开
    MACHINE LEARNING FOR IDENTIFYING IDLE SESSIONS 审中-公开

    公开(公告)号：US20240121311A1

    公开(公告)日：2024-04-11

    申请号：US17962078

    申请日：2022-10-07

    Applicant: Microsoft Technology Licensing, LLC

    Inventor： Prerana Dharmesh GAMBHIR ,  Sharena Meena PARI-MONASCH ,  Khoa Dang NGUYEN ,  Yiming SHI ,  Yongchang DONG

    IPC: H04L67/143 ,  H04L67/54

    CPC classification number: H04L67/143 ,  H04L67/54

    Abstract: Systems and methods for identifying and evicting idle sessions include training a machine learning model as a session classifying model to learn rules for classifying active sessions between clients and the cloud-based service. The session classifying model is trained to receive a plurality of parameters pertaining to the document associated with an active session as input and to apply the rules to the plurality of parameters to determine a classification for the active session and to provide an output indicative of the classification for the active session. The session classifying model is then utilized in the cloud-based service to classify the active sessions. The active sessions classified as idle sessions may then be evicted from the cloud-based service.

    SPONSORED ACCESS TO MULTI-ITEM DOCUMENT BUNDLES
    2.
    发明公开
    SPONSORED ACCESS TO MULTI-ITEM DOCUMENT BUNDLES 审中-公开

    公开(公告)号：US20240078340A1

    公开(公告)日：2024-03-07

    申请号：US17939687

    申请日：2022-09-07

    Applicant: Microsoft Technology Licensing, LLC

    Inventor： Stephen Lincoln RICE ,  Sean Lamont Grant LIVINGSTON ,  Shiguang DONG ,  Zhenguang CHEN ,  Vladimir FESICH ,  Vijay SHARMA

    IPC: G06F21/62

    CPC classification number: G06F21/6272 ,  G06F2221/2141 ,  G06F2221/2145

    Abstract: A data processing system implements detecting that a first user device associated with a first user has added a first sponsored content item to a host electronic document by adding a first reference to the first sponsored content item to the host electronic document. The first sponsored content item is stored separately in a memory of a cloud-based service from the host electronic document. The data processing system further implements determining that the first user has permission to share the first sponsored content item with other users of the cloud-based service, generating sponsor information to associate the sponsored content item with the host electronic document to permit users having access to the host electronic document to access the sponsored content item, and storing the sponsor information in a sponsored access datastore.

    VIEW AUGMENTATION IN MULTISCREEN ENVIRONMENT
    3.
    发明申请
    VIEW AUGMENTATION IN MULTISCREEN ENVIRONMENT 审中-公开

    公开(公告)号：US20190272186A1

    公开(公告)日：2019-09-05

    申请号：US15912369

    申请日：2018-03-05

    Applicant: Microsoft Technology Licensing, LLC

    Inventor： David GROCHOCKI, Jr. ,  Ana Lilia OTERO DIAZ ,  Kshitij SETHI ,  Jeff WEST ,  Felix ANDREW

    IPC: G06F9/451 ,  G06F9/46 ,  G06F3/14

    Abstract: The disclosed technology is generally directed to multiscreen environments. In one example of the technology, a registry is maintained. The registry includes a plurality of context schemas. Each context schema in the plurality of context schemas includes a context schema input and a context schema output. Context data is received from a first application. The context data includes at least one of a context schema input that is associated with view augmentation in a multiscreen environment or a context schema output that is associated with view augmentation in a multiscreen environment. It is determined whether the context data is valid based, at least in part, on the registry. In response to determining that the context data is valid, the context data is forwarded to at least a second application. The second application is separate from the first application.

    PRINTED DE-COUPLING PLANE FOR PRINTED CIRCUIT BOARDS
    4.
    发明申请
    PRINTED DE-COUPLING PLANE FOR PRINTED CIRCUIT BOARDS 有权

    公开(公告)号：US20250168968A1

    公开(公告)日：2025-05-22

    申请号：US18835151

    申请日：2022-02-03

    Applicant: Microsoft Technology Licensing, LLC

    Inventor： Chee Kiong FONG ,  Ying LIANG ,  Geoffrey Jason SHEW ,  Michael Roy VOLKMAN ,  Suet Fong TIN ,  Marshall Thomas DEPUE

    IPC: H05K1/02 ,  H05K1/09 ,  H05K1/18 ,  H05K3/46

    Abstract: Printed circuit boards (PCBs) are a fundamental component used in nearly all electronics. PCBs provide electrical connections and mechanical support to electronic components and are generally made of copper layers laminated onto, though, and/or between one or more non-conductive substrate layers. Depending on the circuit complexity and performance requirements, multiple copper layers may be utilized in a singular PCB. Beyond two copper layers, the material cost and environmental impact of adding additional copper layers to a PCB increases dramatically. Multiple internal conductive layers (e.g., copper clad laminates or CCL's) within a PCB are created using an energy intensive process, requiring substantial amounts of water and chemicals. Significant environmental savings and PCB manufacturing cost reductions can be achieved if a PCB design can minimize the number of internal conductive layers, including using one or more printed de-coupling layers in place of laminated copper layers.

    FACILITATING COLLECTION OF EVENTS DETECTED BY RADIO ACCESS NETWORK COMPONENTS
    5.
    发明申请
    FACILITATING COLLECTION OF EVENTS DETECTED BY RADIO ACCESS NETWORK COMPONENTS 有权

    公开(公告)号：US20250168672A1

    公开(公告)日：2025-05-22

    申请号：US19029672

    申请日：2025-01-17

    Applicant: Microsoft Technology Licensing, LLC

    Inventor： Bhushan Mangesh KANEKAR ,  Prateek MITTAL

    IPC: H04W24/08 ,  H04L67/10

    Abstract: The present disclosure relates to systems, methods, and computer-readable media for collecting operational data across a plurality of radio access network (RAN) components. For example, the systems described herein can identify data signals that are tracked by one or more RAN components. Based on these data signals, the systems can define any number of network events that may be tracked by event tracking agents that are deployed on each of the RAN component(s). The RAN components may then provide a stream of event instances to the systems for collecting, analyzing, and otherwise utilizing the network event data that is locally tracked by the respective RAN components.

    CONTENT ADAPTIVE DEBLOCKING DURING VIDEO ENCODING AND DECODING
    6.
    发明申请
    CONTENT ADAPTIVE DEBLOCKING DURING VIDEO ENCODING AND DECODING 有权

    公开(公告)号：US20250168387A1

    公开(公告)日：2025-05-22

    申请号：US19025888

    申请日：2025-01-16

    Applicant: Microsoft Technology Licensing, LLC

    Inventor： Xiaoyan Sun ,  Zhiwei Xiong ,  Feng Wu

    IPC: H04N19/51 ,  H04N19/117 ,  H04N19/14 ,  H04N19/176 ,  H04N19/80 ,  H04N19/86

    Abstract: Disclosed herein are exemplary embodiments of methods, apparatus, and systems for performing content-adaptive deblocking to improve the visual quality of video images compressed using block-based motion-predictive video coding. For instance, in certain embodiments of the disclosed technology, edge information is obtained using global orientation energy edge detection (“OEED”) techniques on an initially deblocked image. OEED detection can provide a robust partition of local directional features (“LDFs”). For a local directional feature detected in the partition, a directional deblocking filter having an orientation corresponding to the orientation of the LDF can be used. The selected filter can have a filter orientation and activation thresholds that better preserve image details while reducing blocking artifacts. In certain embodiments, for a consecutive non-LDF region, extra smoothing can be imposed to suppress the visually severe blocking artifacts.

    DETECTION OF MALICIOUS ACTIVITY
    7.
    发明申请
    DETECTION OF MALICIOUS ACTIVITY 有权

    公开(公告)号：US20250168183A1

    公开(公告)日：2025-05-22

    申请号：US18868065

    申请日：2023-06-14

    Applicant: Microsoft Technology Licensing, LLC

    Inventor： Amir BEN AMI ,  Amos Avraham RIMON ,  Idan BRUSILOVSKY ,  Danielle POLEG

    IPC: H04L9/40

    Abstract: A method of detecting anomalous events indicative of malicious activity is described. The method comprises receiving a log of an event comprising a plurality of values, the plurality of values comprising known values corresponding to each of a plurality of attributes of the event and generating a masked log by masking a value in the received log, the masked value corresponding to one of the attributes. The method further comprises, based on the masked log and a trained machine learning model, generating a distribution of probabilities for possible values of the masked value, wherein the trained machine learning model is based on a plurality of masked logs of events, and determining that the event is an anomalous event based on a comparison of the known value of the masked value and the distribution of probabilities.

    ADVERSARIAL TRAINING OF MACHINE LEARNING MODELS
    10.
    发明申请
    ADVERSARIAL TRAINING OF MACHINE LEARNING MODELS 有权

    公开(公告)号：US20250165792A1

    公开(公告)日：2025-05-22

    申请号：US19034250

    申请日：2025-01-22

    Applicant: Microsoft Technology Licensing, LLC

    Inventor： Xiaodong LIU ,  Jianfeng GAO ,  Pengcheng HE ,  Weizhu CHEN

    IPC: G06N3/088 ,  G06N3/045

    Abstract: This document relates to training of machine learning models such as neural networks. One example method involves providing a machine learning model having one or more layers and associated parameters and performing a pretraining stage on the parameters of the machine learning model to obtain pretrained parameters. The example method also involves performing a tuning stage on the machine learning model by using labeled training samples to tune the pretrained parameters. The tuning stage can include performing noise adjustment of the labeled training examples to obtain noise-adjusted training samples. The tuning stage can also include adjusting the pretrained parameters based at least on the labeled training examples and the noise-adjusted training examples to obtain adapted parameters. The example method can also include outputting a tuned machine learning model having the adapted parameters.

Patent Agency Ranking