Google Inc.

X Development LLC

Ariel Gordon

Anelia Angelova

Michael Joseph Quinlan

Bo Chen

Guy Satat

Sean Kirmani

Elad Eban

Ofir Nachum

Ehud Benyamin Rivlin

Balakrishnan Varadarajan

Apostol Ivanov Natsev

Zhao Chen

Soeren Pirk

Vincent Michael Casser

Yao Lu

Joonseok Lee

Anthony Brohan

Seong Jae Hwang

Jan Dlabal

Methods, systems, and apparatus, including computer programs encoded on a computer storage medium, for training neural networks. In one aspect, a system includes a neural network shrinking engine that is configured to receive a neural network being trained and generate a reduced neural network by a shrinking process. The shrinking process includes training the neural network based on a shrinking engine loss function that includes terms penalizing active neurons of the neural network and removing inactive neurons from the neural network. The system includes a neural network expansion engine that is configured to receive the neural network being trained and generate an expanded neural network by an expansion process including adding new neurons to the neural network and training the neural network based on an expanding engine loss function. The system includes a training subsystem that generates reduced neural networks and expanded neural networks.

Learning neural network structure

A method includes receiving a first depth map that includes a plurality of first pixel depths and a second depth map that includes a plurality of second pixel depths. The first depth map corresponds to a reference depth scale and the second depth map corresponds to a relative depth scale. The method includes aligning the second pixel depths with the first pixel depths. The method includes transforming the aligned region of the second pixel depths such that transformed second edge pixel depths of the aligned region are coextensive with first edge pixel depths surrounding the corresponding region of the first pixel depths. The method includes generating a third depth map. The third depth map includes a first region corresponding to the first pixel depths and a second region corresponding to the transformed and aligned region of the second pixel depths.

Fusing multiple depth sensing modalities

Methods, systems, and apparatus, including computer programs encoded on computer storage media, for training a neural network using consistency measures. One of the methods includes processing a particular training example from a mediator training data set using a first neural network to generate a first output for a first machine learning task; processing the particular training example in the mediator training data set using each of one or more second neural networks, wherein each second neural network is configured to generate a second output for a respective second machine learning task; determining, for each second machine learning task, a consistency target output for the first machine learning task; determining, for each second machine learning task, an error between the first output and the consistency target output corresponding to the second machine learning task; and generating a parameter update for the first neural network from the determined errors.

Training neural networks using consistency measures

A MapReduce-based training framework exploits both data parallelism and model parallelism to scale training of complex models. Particular model architectures facilitate and benefit from use of such training framework. As one example, a machine-learned model can include a shared feature extraction portion configured to receive and process a data input to produce an intermediate feature representation and a plurality of prediction heads that are configured to receive and process the intermediate feature representation to respectively produce a plurality of predictions. For example, the data input can be a video and the plurality of predictions can be a plurality of classifications for content of the video (e.g., relative to a plurality of classes).

Framework for training machine-learned models on extremely large datasets

Methods and systems are provided that use images to determine lighting information for an object. A computing device can receive an image of the object. For a pixel of the image, the computing device can: apply a first lighting model to determine a first estimate of a bi-directional lighting function (BRDF) for the object at the pixel, apply a second lighting model to determine a second estimate of the BRDF for the object at the pixel, determine a third estimate of the BRDF based on the first and second estimates, and store the third estimate of the BRDF in lighting-storage data. The computing device can provide the lighting-storage data. The BRDF can utilize a number of lighting parameters, such as a normal vector and albedo, reflectivity, and roughness values.

Extraction and representation of three-dimensional (3D) and bidirectional reflectance distribution function (BRDF) parameters from lighted image sequences

Growing community of inventors

Mountain View, CA, United States of America

Ariel Gordon