Google Inc.

Lucent Technologies Inc.

At+t Corp.

Corinna Cortes

Sergey Ioffe

Ameesh Makadia

Hartwig Adam

Teresa Ko

Troy Chinen

Nemanja Petrovic

Sebastian Pueblas

Michael Dennis Riley

Lawrence David Jackel

Daryl Pregibon

Jay Yagnik

Sanjiv Kumar

Theodore Donald Johnson

Gideon S Mann

Ming Zhao

Jeremy Hylton

Ana Radovanovic

Wan-Ping Chiang

Vladimir Vapnik

Jocelyn Miller

Christopher C Pennock

Christopher T Volinsky

William H DuMouchel

William Lee

David Tussey

Methods, systems, and apparatus, including computer programs encoded on computer storage media, for processing inputs using a neural network system that includes a batch normalization layer. One of the methods includes receiving a respective first layer output for each training example in the batch; computing a plurality of normalization statistics for the batch from the first layer outputs; normalizing each component of each first layer output using the normalization statistics to generate a respective normalized layer output for each training example in the batch; generating a respective batch normalization layer output for each of the training examples from the normalized layer outputs; and providing the batch normalization layer output as an input to the second neural network layer.

Batch normalization layers

Methods, systems, and apparatus, including computer programs encoded on computer storage media, for processing images or features of images using an image classification system that includes a batch normalization layer. One of the systems includes a convolutional neural network configured to receive an input comprising an image or image features of the image and to generate a network output that includes respective scores for each object category in a set of object categories, the score for each object category representing a likelihood that that the image contains an image of an object belonging to the category, and the convolutional neural network comprising: a plurality of neural network layers, the plurality of neural network layers comprising a first convolutional neural network layer and a second neural network layer; and a batch normalization layer between the first convolutional neural network layer and the second neural network layer.

Image classification using batch normalization layers

Techniques for providing image search templates are provided. An image search template may be associated with an image search query to aid the user in capturing an image that will be appropriate for processing the search query. The template may be displayed as an overlay during an image capturing process to indicate an appropriate image capturing pose, range, angle, or other view characteristics that may provide more accurate search results. The template may also be used in the image search query to segment the image and identify features relevant to the search query. Images in an image database may be clustered using characteristics of the images or metadata associated with the images in order to establish groups of images from which templates may be derived. The generated templates may be provided to users to assist in capturing images to be used as search engine queries.

Providing image search templates

A statistical model may be created that relates human ratings of documents to objective signals generated from the documents, search queries, and/or other information (e.g., query logs). The model can then be used to predict human ratings/rankings for new documents/search query pairs. These predicted ratings can be used to, for example, refine rankings from a search engine or assist in evaluating or monitoring the efficacy of a search engine system.

Prediction of human ratings or rankings of information retrieval quality

A computer-implemented method for defining a segment based on interaction proneness includes receiving online activity data that specifies instances of presentation for one or more content items, and instances of user interaction detected for any of the content items. The method includes training at least one predictive model on the online activity data, the predictive model trained to predict interaction proneness based on one or more characteristics associated with the instances of user interaction. The method includes identifying, using the predictive model, at least one of the characteristics as being associated with the interaction proneness. The method includes generating at least one segment definition that takes into account the identified characteristic.

Defining a segment based on interaction proneness

Systems and methods relating to website quality rating are disclosed. Websites are rated, relationships between ratings and website signals are identified, models are generated and modeled ratings are assigned to unrated websites by applying the models to the website signals of the unrated websites.

Website quality signal generation

A video demographics analysis system selects a training set of videos to use to correlate viewer demographics and video content data. The video demographics analysis system extracts demographic data from viewer profiles related to videos in the training set and creates a set of demographic distributions, and also extracts video data from videos in the training set. The video demographics analysis system correlates the viewer demographics with the video data of videos viewed by that viewer. Using the prediction model produced by the machine learning process, a new video about which there is no a priori knowledge can be associated with a predicted demographic distribution specifying probabilities of the video appealing to different types of people within a given demographic category, such as people of different ages within an age demographic category.

Video content analysis for automatic demographics recognition of users and videos

Growing community of inventors

New York, NY, United States of America

Corinna Cortes