公开(公告)号：US11257217B2

公开(公告)日：2022-02-22

申请号：US16761381

申请日：2018-11-20

Applicant: Google LLC

Inventor： Jonathon Shlens ,  Niruban Maheswaranathan ,  David Sussillo

IPC: G06T7/11 ,  G06N3/04 ,  G06N3/08

Abstract: A method for generating a segmentation of an image that assigns each pixel to a respective segmentation category from a set of segmentation categories is described. The method includes obtaining features of the image, the image including a plurality of pixels. For each of one or more time steps starting from an initial time step and continuing until a final time step, the method includes generating a network input from the features of the image and a current segmentation output as of the time step, processing the network input using a convolutional recurrent neural network to generate an intermediate segmentation output for the time step, and generating an updated segmentation output for the time step from the intermediate segmentation output for the time step and the current segmentation output as of the time step. The method includes generating a final segmentation of the image from the updated segmentation output.

公开(公告)号：US20220391706A1

公开(公告)日：2022-12-08

申请号：US17831338

申请日：2022-06-02

Applicant: Google LLC

Inventor： Luke Shekerjian Metz ,  Christian Daniel Freeman ,  Jascha Narain Sohl-Dickstein ,  Niruban Maheswaranathan ,  James Michael Harrison

IPC: G06N3/08

Abstract: Methods, systems, and apparatus, including computer programs encoded on computer storage media, for training neural networks using learned optimizers. One of method is for training a neural network layer comprising a plurality of network parameters having a plurality of dimensions each having a plurality of indices, the method comprising: maintaining a set of values corresponding to respective sets of indices of each dimension, each value representing a measure of central tendency of past gradients of the network parameters having an index in the dimension that is in the set of indices; performing a training step to obtain a new gradient for each network parameter; updating each set of values using the new gradients; and for each network parameter: generating an input from the updated sets of values; processing the input using an optimizer neural network to generate an output defining an update for the network parameter; and applying the update.

公开(公告)号：US20220092429A1

公开(公告)日：2022-03-24

申请号：US17481160

申请日：2021-09-21

Applicant: Google LLC

Inventor： Luke Shekerjian Metz ,  Niruban Maheswaranathan ,  Christian Daniel Freeman ,  Benjamin Poole ,  Jascha Narain Sohl-Dickstein

IPC: G06N3/08

Abstract: Methods, systems, and apparatus, including computer programs encoded on computer storage media, for training a neural network. One of the methods includes performing, using a plurality of training examples, a training step to obtain respective gradients of a loss function with respect to each of the parameters in the parameter tensors; obtaining a validation loss for a plurality of validation examples that are different from the plurality of training examples generating an optimizer input from at least the respective gradients and the validation loss; processing the optimizer input using an optimizer neural network to generate an output defining a respective update for each of the parameters in the parameter tensors of the neural network; and for each of the parameters in the parameter tensors, applying the respective update to a current value of the parameter to generate an updated value for the parameter.