-
公开(公告)号：US09390382B2
公开(公告)日：2016-07-12
申请号：US14142970
申请日：2013-12-30
Applicant: Google Inc.
Inventor： Yoram Singer , Tal Shaked , Tushar Deepak Chandra , Tze Way Eugene Ie , James Vincent McFadden , Jeremiah Harmsen , Kristen Riedt LeFevre
IPC: G06N99/00
CPC classification number: G06N99/005
Abstract: Systems and techniques are disclosed for training a machine learning model based on one or more regularization penalties associated with one or more features. A template having a lower regularization penalty may be given preference over a template having a higher regularization penalty. A regularization penalty may be determined based on domain knowledge. A restrictive regularization penalty may be assigned to a template based on determining that a template occurrence is below a stability threshold and may be modified if the template occurrence meets or exceeds the stability threshold.
Abstract translation: 公开了用于基于与一个或多个特征相关联的一个或多个正则化惩罚来训练机器学习模型的系统和技术。 具有较低正则化罚分的模板可以优先于具有较高正则化惩罚的模板。 正规化惩罚可以根据领域知识来确定。 基于确定模板出现低于稳定性阈值，可以将限制性正则化惩罚分配给模板，并且如果模板发生满足或超过稳定性阈值，则可以对模板进行修改。
-
公开(公告)号：US20170300814A1
公开(公告)日：2017-10-19
申请号：US15394668
申请日：2016-12-29
Applicant: Google Inc.
Inventor： Tal Shaked , Rohan Anil , Hrishikesh Balkrishna Aradhye , Mustafa Ispir , Glen Anderson , Wei Chai , Mehmet Levent Koc , Jeremiah Harmsen , Xiaobing Liu , Gregory Sean Corrado , Tushar Deepak Chandra , Heng-Tze Cheng
CPC classification number: G06N3/08 , G06N3/0454 , G06N3/0472 , G06N3/084
Abstract: A system includes one or more computers and one or more storage devices storing instructions that when executed by the one or more computers cause the computers to implement a combined machine learning model for processing an input including multiple features to generate a predicted output for the machine learning input. The combined model includes: a deep machine learning model configured to process the features to generate a deep model output; a wide machine learning model configured to process the features to generate a wide model output; and a combining layer configured to process the deep model output generated by the deep machine learning model and the wide model output generated by the wide machine learning model to generate the predicted output, in which the deep model and the wide model have been trained jointly on training data to generate the deep model output and the wide model output.
-
公开(公告)号：US20170213252A1
公开(公告)日：2017-07-27
申请号：US15431000
申请日：2017-02-13
Applicant: Google Inc.
Inventor： Terrence Rohan , Tomasz J. Tunguz-Zawislak , Jeremiah Harmsen , Sverre Sundsdal , Thomas M. Annau , Megan Nance , Mayur Dhondu Datar , Julie Tung , Bahman Rabii , Jason C. Miller , Michael Hochberg , Andres S. Perez-Bergquist
CPC classification number: G06Q30/0269 , G06Q30/02 , G06Q30/0273 , G06Q30/0277 , G06Q50/01
Abstract: The subject matter of this document generally relates to reducing noise in aggregated data using frequency analysis. In some implementations, a system for reducing data noise using frequency analysis includes a data storage device that stores content and a network association processor in data communication with the data storage device. The network association processor aggregates, for a given group, content of one or more additional groups that each have overlapping members with the given group. The network association processor reduces noise in the aggregated content of the one or more additional groups using frequency analysis by determining, for each portion of content in the aggregated content, a frequency of occurrence of the portion of content within the aggregated content and filtering, from the aggregated content, each portion of content that has a frequency of occurrence that is less than a threshold.
-
Patent Agency Ranking
公开(公告)号：US20170039483A1
公开(公告)日：2017-02-09
申请号：US14820751
申请日：2015-08-07
Applicant: Google Inc.
Inventor： Heng-Tze Cheng , Jeremiah Harmsen , Alexandre Tachard Passos , David Edgar Lluncor , Shahar Jamshy , Tal Shaked , Tushar Deepak Chandra
CPC classification number: G06N99/005 , G06F17/30477 , G06F17/30864 , G06N5/04
Abstract: Methods, systems, and apparatus, including computer programs encoded on a computer storage medium, for training a factorization model to learning features of model inputs of a trained model such that the factorization model is predictive of outcome for which the machine learned model is trained.
Abstract translation: 方法，系统和装置，包括在计算机存储介质上编码的计算机程序，用于训练因子分解模型以学习训练模型的模型输入的特征，使得因式分解模型预测机器学习模型被训练的结果。
-
公开(公告)号：US20150317357A1
公开(公告)日：2015-11-05
申请号：US14268049
申请日：2014-05-02
Applicant: Google Inc.
Inventor： Jeremiah Harmsen , Tushar Deepak Chandra , Marcus Fontoura
CPC classification number: G06F17/30424 , G06F17/30321 , G06F17/30622 , G06N5/025 , G06N99/005
Abstract: Systems and techniques are disclosed for generating entries for a searchable index based on rules generated by one or more machine-learned models. The index entries can include one or more tokens correlated with an outcome and an outcome probability. A subset of tokens can be identified based on the characteristics of an event. The index may be searched for outcomes and their respective probabilities that correspond to tokens that are similar to or match the subset of tokens based on the event.
Abstract translation: 公开了用于基于由一个或多个机器学习模型生成的规则为可搜索索引生成条目的系统和技术。 索引条目可以包括与结果相关联的一个或多个令牌和结果概率。 可以基于事件的特征来识别令牌子集。 搜索索引可以根据事件搜索与令牌子集相似或匹配的令牌对应的结果及其各自的概率。
-
公开(公告)号：US20170286864A1
公开(公告)日：2017-10-05
申请号：US15091381
申请日：2016-04-05
Applicant: Google Inc.
Inventor： Noah Fiedel , Christopher Olston , Jeremiah Harmsen
IPC: G06N99/00
Abstract: Methods, systems, and apparatus, including computer programs encoded on computer storage media, for batching inputs to machine learning models. One of the methods includes receiving a stream of requests, each request identifying a respective input for processing by a first machine learning model; adding the respective input from each request to a first queue of inputs for processing by the first machine learning model; determining, at a first time, that a count of inputs in the first queue as of the first time equals or exceeds a maximum batch size and, in response: generating a first batched input from the inputs in the queue as of the first time so that a count of inputs in the first batched input equals the maximum batch size, and providing the first batched input for processing by the first machine learning model.
-
公开(公告)号：US20150186794A1
公开(公告)日：2015-07-02
申请号：US14142970
申请日：2013-12-30
Applicant: Google Inc.
Inventor： Yoram Singer , Tal Shaked , Tushar Deepak Chandra , Tze Way Eugene Ie , James Vincent McFadden , Jeremiah Harmsen , Kristen Riedt LeFevre
IPC: G06N99/00
CPC classification number: G06N99/005
Abstract: Systems and techniques are disclosed for training a machine learning model based on one or more regularization penalties associated with one or more features. A template having a lower regularization penalty may be given preference over a template having a higher regularization penalty. A regularization penalty may be determined based on domain knowledge. A restrictive regularization penalty may be assigned to a template based on determining that a template occurrence is below a stability threshold and may be modified if the template occurrence meets or exceeds the stability threshold.
Abstract translation: 公开了用于基于与一个或多个特征相关联的一个或多个正则化惩罚来训练机器学习模型的系统和技术。 具有较低正则化罚分的模板可以优先于具有较高正则化惩罚的模板。 正规化惩罚可以根据领域知识来确定。 基于确定模板出现低于稳定性阈值，可以将限制性正则化惩罚分配给模板，并且如果模板发生满足或超过稳定性阈值，则可以对模板进行修改。
-
公开(公告)号：US08744911B2
公开(公告)日：2014-06-03
申请号：US13888236
申请日：2013-05-06
Applicant: Google Inc.
Inventor： Terrence Rohan , Tomasz J. Tunguz-Zawislak , Scott G. Sheffer , Jeremiah Harmsen
IPC: G06Q30/00
CPC classification number: G06Q30/0269 , G06Q30/02 , G06Q30/0226 , G06Q30/0241 , G06Q50/01
Abstract: A computer-implemented method for displaying advertisements to members of a network comprises identifying one or more communities of members, identifying one or more influencers in the one or more communities, and placing one or more advertisements at the profiles of one or more members in the identified one or more communities.
Abstract translation: 用于向网络成员显示广告的计算机实现的方法包括识别一个或多个成员社区，识别所述一个或多个社区中的一个或多个影响者，以及将一个或多个广告放置在所述一个或多个成员的简档中。 确定了一个或多个社区。
-
-
-
-
-
-
-
Search Results
8
records
(took 0.013 seconds)
