公开(公告)号：US20180114108A1

公开(公告)日：2018-04-26

申请号：US15787615

申请日：2017-10-18

Applicant: Google Inc.

Inventor： Ni Lao ,  Lukasz Mieczyslaw Kaiser ,  Nitin Gupta ,  Afroz Mohiuddin ,  Preyas Popat

IPC: G06N3/00 ,  G06N3/04

CPC classification number: G06N3/006 ,  G06N3/0445 ,  G06N3/0454 ,  G06N3/084 ,  G06N5/04

Abstract: Methods, systems, and apparatus, including computer programs encoded on computer storage media, for identifying answers to questions using neural networks. One of the methods includes receiving an input text passage and an input question string; processing the input text passage using an encoder neural network to generate a respective encoded representation for each passage token in the input text passage; at each time step: processing a decoder input using a decoder neural network to update the internal state of the decoder neural network; and processing the respective encoded representations and a preceding output of the decoder neural network using a matching vector neural network to generate a matching vector for the time step; and generating an answer score that indicates how well the input text passage answers a question posed by the input question string.

公开(公告)号：US11093813B2

公开(公告)日：2021-08-17

申请号：US15787615

申请日：2017-10-18

Applicant: Google Inc.

Inventor： Ni Lao ,  Lukasz Mieczyslaw Kaiser ,  Nitin Gupta ,  Afroz Mohiuddin ,  Preyas Popat

IPC: G06N5/04 ,  G06N3/00 ,  G06N3/04 ,  G06N3/08

Abstract: Methods, systems, and apparatus, including computer programs encoded on computer storage media, for identifying answers to questions using neural networks. One of the methods includes receiving an input text passage and an input question string; processing the input text passage using an encoder neural network to generate a respective encoded representation for each passage token in the input text passage; at each time step: processing a decoder input using a decoder neural network to update the internal state of the decoder neural network; and processing the respective encoded representations and a preceding output of the decoder neural network using a matching vector neural network to generate a matching vector for the time step; and generating an answer score that indicates how well the input text passage answers a question posed by the input question string.

公开(公告)号：US10810193B1

公开(公告)日：2020-10-20

申请号：US13801598

申请日：2013-03-13

Applicant: Google Inc.

Inventor： Amarnag Subramanya ,  Fernando Pereira ,  Ni Lao ,  John Blitzer ,  Rahul Gupta

IPC: G06F16/245

Abstract: Implementations include systems and methods for querying a data graph. An example method includes receiving a machine learning module trained to produce a model with multiple features for a query, each feature representing a path in a data graph. The method also includes receiving a search query that includes a first search term, mapping the search query to the query, and mapping the first search term to a first entity in the data graph. The method may also include identifying a second entity in the data graph using the first entity and at least one of the multiple weighted features, and providing information relating to the second entity in a response to the search query. Some implementations may also include training the machine learning module by, for example, generating positive and negative training examples from an answer to a query.