DISTRIBUTED DEEP LEARNING USING A DISTRIBUTED DEEP NEURAL NETWORK

    公开(公告)号:US20180307979A1

    公开(公告)日:2018-10-25

    申请号:US15491950

    申请日:2017-04-19

    CPC classification number: G06N3/08 G06N5/04 H04N7/183

    Abstract: Apparatus and associated methods relate to training a neural network on a first host system, sending the neural network to a second host system, training the neural network by the second host system based on data private to the second host system, and employing the neural network to filter events sent to the first host system. In an illustrative example, the first host system may be a server having a central repository including trained neural networks and historical data, and the second host system may be a remote server having a data source private to the remote server. The private remote data source may be a camera. In some examples, events may be filtered as a function of a prediction of error in the neural network. Various examples may advantageously provide remote intelligent filtering. For example, remote data may remain private while adaptively filtering events to the central server.

    Automatic threat detection based on video frame delta information in compressed video streams

    公开(公告)号:US11074791B2

    公开(公告)日:2021-07-27

    申请号:US16529907

    申请日:2019-08-02

    Abstract: Apparatus and associated methods relate to identifying objects of interest and detecting motion to automatically detect a security threat as a function of video frame delta information received from a video encoder. In an illustrative example, the video encoder may be an H.264 encoder onboard a video camera. A cloud server may receive the video frame delta information in a compressed video stream from the camera. Threats may be detected by the cloud server processing the video frame delta information in the compressed video stream, without decompression, to identify objects and detect motion. The cloud server may employ artificial intelligence techniques to enhance threat detection by the cloud server. Various examples may advantageously provide increased capacity of a computer tasked with detecting security breaches, due to the significant reduction in the amount of data to be processed, relative to threat detection based on processing uncompressed video streams.

    Contextual analysis and control of content item selection

    公开(公告)号:US10748159B1

    公开(公告)日:2020-08-18

    申请号:US13179365

    申请日:2011-07-08

    Abstract: Techniques are described related to selecting content items, such as by enabling user analysis and control of product-related content items selected for display to users. The content items may include advertisements or other promotional materials, and the selecting of the content items may be performed as part of determining particular promotional materials to display or otherwise present to particular users in particular situations. User analysis and control of selected content items that are displayed on a target electronic site may be enabled by providing, as part of the target electronic site, additional selection-related functionality whose availability is restricted to one or more authorized users—for example, such additional restricted access information and user-selectable controls may be provided on a version of a Web page of an online retailer to enable the retailer to analyze and influence future content items selected for display on the online retailer's Web page(s).

    GENERATION OF PRODUCT RECOMMENDATIONS USING A DYNAMICALLY SELECTED STRATEGY
    5.
    发明申请
    GENERATION OF PRODUCT RECOMMENDATIONS USING A DYNAMICALLY SELECTED STRATEGY 审中-公开
    使用动态选择的策略产生产品建议

    公开(公告)号:US20120310771A1

    公开(公告)日:2012-12-06

    申请号:US13584611

    申请日:2012-08-13

    CPC classification number: G06Q30/0631 G06Q30/02 G06Q30/0601 G06Q30/0643

    Abstract: Techniques are described for dynamically generating recommendations for users, such as for products and other items. In at least some situations, the techniques include using multiple recommendation strategies, such as by aggregating recommendation results from multiple different recommendation strategies. Such recommendation strategies may have various forms, and may be based at least in part on data regarding prior interactions of numerous users with numerous items. In addition, information about current selections of a particular user may be gathered based at least in part on providing a GUI (“graphical user interface”) for display to the user that includes selectable information about numerous recommended items, and dynamically updating the displayed GUI with newly generated recommendations of items as the user makes selections of particular displayed recommended items (e.g., newly generated recommendations that are similar to the selected items in one or more manners, or are otherwise related to the selected items).

    Abstract translation: 描述了用于为用户动态生成建议的技术,例如用于产品和其他项目。 在至少一些情况下,这些技术包括使用多个推荐策略,例如通过聚合来自多个不同推荐策略的推荐结果。 这样的推荐策略可以具有各种形式,并且可以至少部分地基于关于具有许多项目的许多用户的先前交互的数据。 另外,关于特定用户的当前选择的信息可以至少部分地基于提供用于向用户显示的GUI(图形用户界面)来收集,其包括关于许多推荐项目的可选择的信息,并且用新的动态地更新所显示的GUI 当用户选择特定显示的推荐项目(例如,以一种或多种方式与所选项目相似的新生成的建议或与所选项目相关的新产生的建议)时产生项目的建议。

    MULTI-STRATEGY GENERATION OF PRODUCT RECOMMENDATIONS
    7.
    发明申请
    MULTI-STRATEGY GENERATION OF PRODUCT RECOMMENDATIONS 有权
    产品推荐的多策略生成

    公开(公告)号:US20100250336A1

    公开(公告)日:2010-09-30

    申请号:US12415896

    申请日:2009-03-31

    CPC classification number: G06Q30/0631 G06Q30/02 G06Q30/0601 G06Q30/0643

    Abstract: Techniques are described for dynamically generating recommendations for users, such as for products and other items. In at least some situations, the techniques include using multiple recommendation strategies, such as by aggregating recommendation results from multiple different recommendation strategies. Such recommendation strategies may have various forms, and may be based at least in part on data regarding prior interactions of numerous users with numerous items. In addition, information about current selections of a particular user may be gathered based at least in part on providing a GUI (“graphical user interface”) for display to the user that includes selectable information about numerous recommended items, and dynamically updating the displayed GUI with newly generated recommendations of items as the user makes selections of particular displayed recommended items (e.g., newly generated recommendations that are similar to the selected items in one or more manners, or are otherwise related to the selected items).

    Abstract translation: 描述了用于为用户动态生成建议的技术,例如用于产品和其他项目。 在至少一些情况下,这些技术包括使用多个推荐策略,例如通过聚合来自多个不同推荐策略的推荐结果。 这样的推荐策略可以具有各种形式,并且可以至少部分地基于关于具有许多项目的许多用户的先前交互的数据。 此外,可以至少部分地基于提供用于向用户显示的GUI(“图形用户界面”)来收集关于特定用户的当前选择的信息,所述GUI(包括图形用户界面)包括关于许多推荐项目的可选择的信息,并且动态地更新所显示的GUI 随着用户选择特定显示的推荐项目(例如,以一种或多种方式类似于所选项目的新生成的建议或者与所选项目相关),新产生的项目建议。

    SYSTEM AND METHOD FOR A SECURITY CHECKPOINT USING RADIO SIGNALS

    公开(公告)号:US20180308331A1

    公开(公告)日:2018-10-25

    申请号:US15492008

    申请日:2017-04-20

    CPC classification number: G08B13/22 G01S5/0284 G07C9/00111 G08B27/005

    Abstract: A security device for monitoring the radio frequency signals generated by mobile phones and similar mobile computing and communication devices. The security device employs an antennae array and computer process that are configured to detect and provide a “fingerprint” for a mobile device based on the unique identifiers contained with the radio and other wireless signals utilized by such mobile device. The “fingerprint” that is obtained can be used to keep track of mobile devices as those devices enter and leave the area of the security device. Moreover, the security device can provide an alert when any new, foreign, or otherwise unrecognized device is within range of the security device and share “fingerprints” and alerts with other security devices in its network.

    Multi-strategy generation of product recommendations
    9.
    发明授权
    Multi-strategy generation of product recommendations 有权
    多策略生成产品推荐

    公开(公告)号:US08244564B2

    公开(公告)日:2012-08-14

    申请号:US12415896

    申请日:2009-03-31

    CPC classification number: G06Q30/0631 G06Q30/02 G06Q30/0601 G06Q30/0643

    Abstract: Techniques are described for dynamically generating recommendations for users, such as for products and other items. In at least some situations, the techniques include using multiple recommendation strategies, such as by aggregating recommendation results from multiple different recommendation strategies. Such recommendation strategies may have various forms, and may be based at least in part on data regarding prior interactions of numerous users with numerous items. In addition, information about current selections of a particular user may be gathered based at least in part on providing a GUI (“graphical user interface”) for display to the user that includes selectable information about numerous recommended items, and dynamically updating the displayed GUI with newly generated recommendations of items as the user makes selections of particular displayed recommended items (e.g., newly generated recommendations that are similar to the selected items in one or more manners, or are otherwise related to the selected items).

    Abstract translation: 描述了用于为用户动态生成建议的技术,例如用于产品和其他项目。 在至少一些情况下,这些技术包括使用多个推荐策略,例如通过聚合来自多个不同推荐策略的推荐结果。 这样的推荐策略可以具有各种形式,并且可以至少部分地基于关于具有许多项目的许多用户的先前交互的数据。 此外,可以至少部分地基于提供用于向用户显示的GUI(“图形用户界面”)来收集关于特定用户的当前选择的信息,所述GUI(包括图形用户界面)包括关于许多推荐项目的可选择的信息,并且动态地更新所显示的GUI 随着用户选择特定显示的推荐项目(例如,以一种或多种方式类似于所选项目的新生成的建议或者与所选项目相关),新产生的项目建议。

    SYSTEM AND METHOD FOR A SECURITY CHECKPOINT USING RADIO SIGNALS

    公开(公告)号:US20190035243A1

    公开(公告)日:2019-01-31

    申请号:US16151654

    申请日:2018-10-04

    Abstract: A security device for monitoring the radio frequency signals generated by mobile phones and similar mobile computing and communication devices. The security device employs an antennae array and computer process that are configured to detect and provide a “fingerprint” for a mobile device based on the unique identifiers contained with the radio and other wireless signals utilized by such mobile device. The “fingerprint” that is obtained can be used to keep track of mobile devices as those devices enter and leave the area of the security device. Moreover, the security device can provide an alert when any new, foreign, or otherwise unrecognized device is within range of the security device and share “fingerprints” and alerts with other security devices in its network.

Patent Agency Ranking