-
公开(公告)号：US20200005191A1
公开(公告)日：2020-01-02
申请号：US16021086
申请日：2018-06-28
IPC分类号： G06N99/00
摘要： An input dataset for training a new machine learning model is received by a processor. For each of a plurality of trained machine learning models, a hash function and a sketch of a training dataset used to train the machine learning model is retrieved. A sketch of the input dataset is computed based on the hash function and the input dataset, along with a distance between the sketch of the training dataset and the sketch of the input dataset. The computed distances of the trained machine learning models are ranked from smallest to largest, and a seed machine learning model for the input dataset is selected from the trained machine learning models based at least in part on the ranking. A training process of the new machine learning model using the selected seed machine learning model and the input dataset is initiated.
-
公开(公告)号：US20220156494A1
公开(公告)日：2022-05-19
申请号：US17098501
申请日：2020-11-16
摘要： Aspects of the invention include generating a combined raster image from point cloud data and reference data describing an original location of a power line. Selecting a set of candidate pixels from the combined raster image describing an updated location of a power line, wherein the selection is based at least in part on a location of pixels in the combined raster image that describe the original location. Detecting pixels from the set of candidate pixels that describe an updated location of a power line. Modifying the combined raster image to reflect the updated location of the power line.
-
公开(公告)号：US20190087720A1
公开(公告)日：2019-03-21
申请号：US15706809
申请日：2017-09-18
摘要： An output time-series of a cell of a neural network is captured. A subset of a set of data points of the output time-series is consolidated into a singular data point. The singular data point is fitted in a data representation to form a quantified aggregated data point. The quantified aggregated data point is included in an intermediate time-series. Using the intermediate time-series as an input at an intermediate layer of the neural network, an anonymized output time-series is produced from the neural network.
-
公开(公告)号：US20230080981A1
公开(公告)日：2023-03-16
申请号：US17447449
申请日：2021-09-13
IPC分类号： G06Q10/06 , G06F16/35 , G06F16/335 , G06F16/332
摘要： In an approach to predictive maintenance explanations based on a user profile, one or more computer processors receive a failure prediction associated with a physical asset of an organization. One or more computer processors receive a work order associated with the failure prediction and an assignment of a first user to the work order. One or more computer processors retrieve a profile associated with the first user. One or more computer processors determine a best match between a taxonomy node of a taxonomy of user expertise associated with the work order and the retrieved profile associated with the first user. Based on the determined best match, one or more computer processors generate an explanation of the failure prediction. One or more computer processors display the explanation to the first user.
-
公开(公告)号：US20210158203A1
公开(公告)日：2021-05-27
申请号：US16691326
申请日：2019-11-21
发明人： RAGHU KIRAN GANTI , MUDHAKAR SRIVATSA , Shreeranjani Srirangamsridharan , Yeon-sup Lim , Linsong Chu
摘要： A computer-implemented method for generating and deploying a reinforced learning model to train a chatbot. The method includes selecting a plurality of conversations, wherein each conversation includes an agent and a user. The method includes identifying, in each of the conversations, a set of turns and on or more topics. The method further includes associating one or more topics to each turn of the set of turns. The method includes, generating a conversation flow for each conversation, wherein the conversation flow identifies a sequence of the topics. The method includes applying an outcome score to each conversation. The method includes creating a reinforced learning (RL) model, wherein the RL model includes a Markov is based on the conversation flow of each conversation and the outcome score of each conversation. The method includes deploying the RL model, wherein the deploying includes sending the RL model to a chatbot.
-
公开(公告)号：US20170310680A1
公开(公告)日：2017-10-26
申请号：US15397946
申请日：2017-01-04
IPC分类号： H04L29/06
CPC分类号： H04L63/105 , H04L63/08
摘要： A method indicates a trustworthiness of data processed in accordance with a processing rule. A first trust weight is assigned to a data item to be processed to provide a weighted data item, the first trust weight representing a level of trust in the data item. A trust value is selected from a set of data trust values, the selected trust value being representative of a determined level of trust in the data item. The selected trust value is defined as the first trust weight which is associated with the data item. The first trust weight is assigned to a processing rule to provide a weighted processing rule, the first trust weight representing a level of trust in the processing rule. The weighted data item is processed in accordance with the weighted processing rule to generate a data output and an indication of a trust level for the data output.
-
公开(公告)号：US20150350891A1
公开(公告)日：2015-12-03
申请号：US14725184
申请日：2015-05-29
摘要： A technique for anonymizing location information of a mobile device includes intercepting, by the mobile device, a request for access to actual location information of the mobile device made by an application. The mobile device redirects the request to a trusted edge server. The trusted edge server obfuscates the location information based on a location of one or more other mobile devices proximal to the mobile device to generate obfuscated location information. The one or more other mobile devices are selected based on a similarity between a user profile associated with the mobile device and user profiles associated with the one or more other mobile devices. The trusted edge server returns the obfuscated location information to the mobile device, which sends the obfuscated location information to the application in place of the requested location information.
摘要翻译： 用于匿名移动设备的位置信息的技术包括由移动设备拦截访问由应用所做的移动设备的实际位置信息的请求。 移动设备将请求重定向到可信边缘服务器。 信任边缘服务器基于位于移动设备附近的一个或多个其他移动设备的位置来模糊位置信息，以产生混淆的位置信息。 基于与移动设备相关联的用户简档和与一个或多个其他移动设备相关联的用户简档之间的相似性来选择一个或多个其他移动设备。 信任边缘服务器将混淆的位置信息返回到移动设备，移动设备将混淆的位置信息发送到应用，而不是所请求的位置信息。
-
公开(公告)号：US20220271992A1
公开(公告)日：2022-08-25
申请号：US17183456
申请日：2021-02-24
摘要： A computer-implemented method and a computer system establish network slices within a physical network having a plurality of network elements. The method includes receiving a request to instantiate a network slice at a network element. The method also includes determining a performance metric of the network element. The method further includes delaying instantiation of the requested network slice within the network element in response to determining that the performance metric of the network element is below a threshold. The method also includes instantiating the requested network slice within the network element in response to determining that the performance metric of the network element is at or above the threshold. Finally, the method includes deactivating the requested network slice in response to determining that the performance metric of the network element is below the threshold at a time subsequent to instantiating the requested network slice.
-
公开(公告)号：US20200342304A1
公开(公告)日：2020-10-29
申请号：US16394006
申请日：2019-04-25
摘要： A method, computer system, and a computer program product for identifying feature importance in deep learning models is provided. Embodiments of the present invention may include building a reconstruction model. Embodiments of the present invention may include intercepting an output of a trained prediction model at a bottleneck layer. Embodiments of the present invention may include processing the output of the trained model using the reconstruction model. Embodiments of the present invention may include identifying a plurality of features based on the reconstruction model.
-
公开(公告)号：US20200293614A1
公开(公告)日：2020-09-17
申请号：US16352358
申请日：2019-03-13
发明人： MUDHAKAR SRIVATSA , RAGHU KIRAN GANTI , Yeon-sup Lim , Shreeranjani Srirangamsridharan , Antara Palit
IPC分类号： G06F17/27
摘要： A parse tree corresponding to a portion of narrative text is constructed. The parse tree includes a data structure representing a syntactic structure of the portion of narrative text as a set of tokens according to a grammar. Using a token in the parse tree as a focus word, a context window comprising a set of words within a specified distance from the focus word is generated, the distance determined according to a number of links of the parse tree separating the focus word and a context word in the set of words. A weight is generated for the focus word and the context word. Using the weight, a first vector representation of a first word is generated, the first word being within a second portion of narrative text.
-
-
-
-
-
-
-
-
-
搜索结果
17 条记录 (耗时 0.063 秒)