公开(公告)号：WO2017027339A1

公开(公告)日：2017-02-16

申请号：PCT/US2016/045681

申请日：2016-08-05

Applicant: GOOGLE INC.

Inventor： RIFKIN, Ryan M. ,  RAMAGE, Daniel

IPC: G10L15/26 ,  H04N5/232 ,  G10L15/22

CPC classification number: H04N5/23203 ,  G06F3/00 ,  G06K9/00355 ,  G06K9/00919 ,  G10L15/22 ,  G10L15/26 ,  G10L2015/223 ,  H04N5/23219

Abstract: The present disclosure relates to a method for controlling a digital photography system. The method includes obtaining, by a device, image data and audio data. The method also includes identifying one or more objects in the image data and obtaining a transcription of the audio data. The method also includes controlling a future operation of the device based at least on the one or more objects identified in the image data, and the transcription of the audio data.

Abstract translation: 本公开涉及一种用于控制数字摄影系统的方法。 该方法包括通过设备获得图像数据和音频数据。 该方法还包括识别图像数据中的一个或多个对象并获得音频数据的转录。 该方法还包括至少基于图像数据中识别的一个或多个对象以及音频数据的转录来控制设备的将来操作。

公开(公告)号：WO2016032777A1

公开(公告)日：2016-03-03

申请号：PCT/US2015/045346

申请日：2015-08-14

Applicant: GOOGLE INC.

Inventor： RAMAGE, Daniel ,  KAHN, Jeremy Gillmor

IPC: G06N3/02

CPC classification number: G06N99/005 ,  H04L67/42

Abstract: Methods, systems, and apparatus, including computer programs encoded on a computer storage medium, for obtaining a global model for a particular activity, the global model derived based on input data representing multiple observations associated with the particular activity performed by a collection of users; determining, using the global model, expected data representing an expected observation associated with the particular activity performed by a particular user; receiving, by a computing device operated by the particular user, particular data representing an actual observation associated with the particular activity performed by the particular user; determining, by the computing device and using (i) the expected data and (ii) the particular data, residual data of the particular user; and deriving a local model of the particular user based on the residual data.

Abstract translation: 方法，系统和装置，包括在计算机存储介质上编码的用于获得用于特定活动的全局模型的计算机程序，所述全局模型基于表示与由用户集合执行的特定活动相关联的多个观察点的输入数据导出; 使用全局模型确定表示与由特定用户执行的特定活动相关联的预期观察值的预期数据; 由特定用户操作的计算设备接收表示与由特定用户执行的特定活动相关联的实际观察的特定数据; 由计算设备确定并使用（i）预期数据和（ii）特定数据，特定用户的剩余数据; 以及基于所述残差数据导出所述特定用户的本地模型。

公开(公告)号：WO2017059388A1

公开(公告)日：2017-04-06

申请号：PCT/US2016/055055

申请日：2016-09-30

Applicant: GOOGLE INC.

Inventor： SHARIFI, Matthew ,  RAMAGE, Daniel ,  PETROU, David

IPC: G06F3/0482 ,  G06F17/30 ,  G06F9/48 ,  G06F9/54

CPC classification number: G06N5/02 ,  G06F3/0481 ,  G06F3/0482 ,  G06F3/0484 ,  G06F3/04895 ,  G06F9/453 ,  G06F17/30424

Abstract: Systems and methods are provided for suggesting actions for selected text based on content displayed on a mobile device. An example method can include converting a selection made via a display device into a query, providing the query to an action suggestion model that is trained to predict an action given a query, each action being associated with a mobile application, receiving one or more predicted actions, and initiating display of the one or more predicted actions on the display device. Another example method can include identifying, from search records, queries where a website is highly ranked, the website being one of a plurality of websites in a mapping of websites to mobile applications. The method can also include generating positive training examples for an action suggestion model from the identified queries, and training the action suggestion model using the positive training examples.

Abstract translation: 系统和方法被提供用于基于在移动设备上显示的内容来建议针对所选文本的动作。 示例性方法可以包括将通过显示设备进行的选择转换成查询，将查询提供给被训练来预测给定查询的动作的动作建议模型，每个动作与移动应用相关联，接收一个或多个预测的 动作，并且在显示装置上开始显示一个或多个预测动作。 另一个示例性方法可以包括从搜索记录中识别网站被高度排名的查询，该网站是将网站映射到移动应用的多个网站之一。 该方法还可以包括从所识别的查询中产生针对动作建议模型的肯定训练示例，以及使用积极训练示例来训练动作建议模型。

公开(公告)号：WO2017112796A1

公开(公告)日：2017-06-29

申请号：PCT/US2016/068083

申请日：2016-12-21

Applicant: GOOGLE INC.

Inventor： GERSHONY, Ori ,  NAZAROV, Sergey ,  DE CASTRO, Rodrigo ,  PALMER, Erika ,  RAMAGE, Daniel ,  RODRIGUEZ, Adam ,  PASCOVICI, Andrei

IPC: H04L12/58

CPC classification number: H04L51/02 ,  H04L51/046 ,  H04L51/063 ,  H04L51/16 ,  H04L51/18 ,  H04L51/20

Abstract: A messaging application may automatically analyze content of one or more messages and/or user information to automatically provide suggestions to a user within a messaging application. The suggestions may automatically incorporate particular non-messaging functionality into the messaging application. The automatic suggestions may suggest one or more appropriate responses to be selected by a user to respond in the messaging application, and/or may automatically send one or more appropriate responses on behalf of a user.

Abstract translation: 消息收发应用可以自动分析一个或多个消息和/或用户信息的内容，以自动向消息收发应用内的用户提供建议。 建议可能会自动将特定的非消息传递功能合并到消息传递应用程序中 自动建议可以建议用户选择一个或多个适当的响应以在消息传递应用中作出响应，和/或可以代表用户自动发送一个或多个适当的响应。

公开(公告)号：WO2017066509A1

公开(公告)日：2017-04-20

申请号：PCT/US2016/056954

申请日：2016-10-14

Applicant: GOOGLE INC.

Inventor： MCMAHAN, Hugh Brendan ,  KONECNY, Jakub ,  MOORE, Eider Brantly ,  RAMAGE, Daniel ,  AGUERA-ARCAS, Blaise H.

IPC: G06F17/30

CPC classification number: G06F17/17 ,  G06F17/11 ,  G06F17/50 ,  G06F2217/04 ,  G06N99/005

Abstract: Systems and methods of determining a global model are provided. In particular, one or more local updates can be received from a plurality of user devices. Each local update can be determined by the respective user device based at least in part on one or more data examples stored on the user device. The one or more data examples stored on the plurality of user devices are distributed on an uneven basis, such that no user device includes a representative sample of the overall distribution of data examples. The local updates can then be aggregated to determine a global model.

Abstract translation: 提供了确定全局模型的系统和方法。 特别地，可以从多个用户设备接收一个或多个本地更新。 每个本地更新可以由相应用户设备至少部分地基于存储在用户设备上的一个或多个数据示例来确定。 存储在多个用户设备上的一个或多个数据示例是不均匀分布的，使得没有用户设备包括数据示例的整体分布的代表性样本。 然后可以汇总本地更新以确定全局模型。

公开(公告)号：WO2014165186A2

公开(公告)日：2014-10-09

申请号：PCT/US2014/024686

申请日：2014-03-12

Applicant: GOOGLE INC.

Inventor： CHUANG, Pi-Chuan ,  RAMAGE, Daniel

IPC: G06F17/30

CPC classification number: G06F17/3053 ,  G06F17/30699 ,  G06F17/30867 ,  G06F17/30876

Abstract: A system stores a table mapping users to attributes, and stores a second table mapping the users to products associated with a source domain. The system determines a set of top scoring products for each of the attributes, and creates, using the top scoring products, a model that is predictive of an activity in a target domain, the target domain being separate from the source domain. The system detects a behavior from a particular user accessing the target domain, and generates a personalized prediction for the particular user based on the model, in response to the detecting the behavior.

Abstract translation: 系统存储将用户映射到属性的表，并存储将用户映射到与源域相关联的产品的第二表。 系统为每个属性确定一组最高评分产品，并使用最高评分产品创建一个预测目标域中活动的模型，目标域与源域分离。 系统检测来自访问目标域的特定用户的行为，并且响应于检测到行为而基于模型为特定用户生成个性化预测。

公开(公告)号：EP3186751A1

公开(公告)日：2017-07-05

申请号：EP15754390.1

申请日：2015-08-14

Applicant: Google Inc.

Inventor： RAMAGE, Daniel ,  KAHN, Jeremy Gillmor

IPC: G06N3/02

CPC classification number: G06N99/005 ,  H04L67/42

Abstract: Methods, systems, and apparatus, including computer programs encoded on a computer storage medium, for obtaining a global model for a particular activity, the global model derived based on input data representing multiple observations associated with the particular activity performed by a collection of users; determining, using the global model, expected data representing an expected observation associated with the particular activity performed by a particular user; receiving, by a computing device operated by the particular user, particular data representing an actual observation associated with the particular activity performed by the particular user; determining, by the computing device and using (i) the expected data and (ii) the particular data, residual data of the particular user; and deriving a local model of the particular user based on the residual data.

Abstract translation: 用于获得特定活动的全局模型的方法，系统和装置，包括编码在计算机存储介质上的计算机程序，所述全局模型基于表示与由用户集合执行的特定活动相关联的多个观察的输入数据而导出; 使用全局模型确定表示与由特定用户执行的特定活动相关联的预期观察的预期数据; 通过由特定用户操作的计算设备接收表示与由特定用户执行的特定活动相关联的实际观察的特定数据; 通过计算设备并且使用（i）预期数据和（ii）特定数据来确定特定用户的残差数据; 以及基于残差数据导出特定用户的局部模型。

公开(公告)号：EP2973026A2

公开(公告)日：2016-01-20

申请号：EP14717308.2

申请日：2014-03-12

Applicant: Google Inc.

Inventor： CHUANG, Pi-Chuan ,  RAMAGE, Daniel

IPC: G06F17/30

CPC classification number: G06F17/3053 ,  G06F17/30699 ,  G06F17/30867 ,  G06F17/30876

Abstract: A system stores a table mapping users to attributes, and stores a second table mapping the users to products associated with a source domain. The system determines a set of top scoring products for each of the attributes, and creates, using the top scoring products, a model that is predictive of an activity in a target domain, the target domain being separate from the source domain. The system detects a behavior from a particular user accessing the target domain, and generates a personalized prediction for the particular user based on the model, in response to the detecting the behavior.