Image classification neural networks

Patent metadata

Field	Value
Publication number	US-2021334605-A1
Application number	US-202117372090-A
Country	US
Kind code	A1
Filing date	Jul 9, 2021
Priority date	Feb 18, 2016
Publication date	Oct 28, 2021
Grant date	—

Abstract

A neural network system that includes: multiple subnetworks that includes: a first subnetwork including multiple first modules, each first module including: a pass-through convolutional layer configured to process the subnetwork input for the first subnetwork to generate a pass-through output; an average pooling stack of neural network layers that collectively processes the subnetwork input for the first subnetwork to generate an average pooling output; a first stack of convolutional neural network layers configured to collectively process the subnetwork input for the first subnetwork to generate a first stack output; a second stack of convolutional neural network layers that are configured to collectively process the subnetwork input for the first subnetwork to generate a second stack output; and a concatenation layer configured to concatenate the pass-through output, the average pooling output, the first stack output, and the second stack output to generate a first module output for the first module.

First claim

What is claimed is:

1 . (canceled)

2 . A system comprising one or more computers and one or more storage devices storing instructions that, when executed by the one or more computers, cause the one or more computers to implement a neural network comprising:
a residual subnetwork comprising a residual module, the residual module comprising:
one or more groups of convolutional neural network layers, each of the one or more groups of convolutional neural network layers being configured to process a subnetwork input for the residual subnetwork to generate a respective group output;
a filter expansion layer that is configured to generate an expanded output by scaling up the dimensionality of each of the group outputs; and
a summing layer configured to generate a summed output from the subnetwork input for the first residual subnetwork and the expanded output.

3 . The neural network system of claim 2 , wherein the residual module further comprises a pass through convolutional layer configured to process the subnetwork input to generate a pass-through output

4 . The neural network system of claim 3 , wherein the filter expansion layer is configured to generate the expanded output by scaling up the dimensionality of each of the group outputs and the pass-through output.

5 . The neural network system of claim 3 , wherein the pass-through convolutional layer is a 1×1 convolutional layer.

6 . The neural network system of claim 3 , wherein the filter expansion layer is configured to receive the pass-through output and the group outputs and to apply a 1×1 convolution to the pass-through output and the group outputs to generate the expanded output.

7 . The neural network system claim 2 , wherein the summing layer is configured to:
sum the subnetwork input for the first subnetwork and the expanded output to generate the summed output.

8 . The neural network system of claim 2 , wherein the summing layer is configured to:
scale the expanded output to generate a scaled expanded output; and sum the subnetwork input for the first subnetwork and the scaled expanded output to generate the summed output.

9 . The neural network system of claim 2 , wherein the residual module further comprises: an activation function layer configured to apply an activation function to the summed output to generate a residual module output for the residual module.

10 . The neural network system of claim 9 , wherein the activation function is a rectified linear unit (Relu) activation function.

11 . The neural network system of claim 2 , wherein the one or more groups of convolutional neural network layers comprises a first stack of convolutional neural network layers.

12 . The neural network system of claim 11 , wherein the first stack of convolutional neural network layers comprises a 1×1 convolutional layer followed by a 1×1 convolutional layer.

13 . The neural network system of claim 11 , wherein the one or more groups of convolutional neural network layers comprises a second stack of convolutional neural network layers.

14 . The neural network system of claim 13 , wherein the second stack of convolutional neural network layers comprises a 1×1 convolutional layer followed by a 3×3 convolutional layer followed by a 3×3 convolutional layer.

15 . The neural network system of claim 9 , wherein the residual subnetwork comprises a plurality of other residual modules and is configured to:
combine the residual module output of the residual module with other residual module outputs of other residual modules to generate a residual subnetwork output for the residual subnetwork.

Assignees

• Google Llc

Inventors

• Vanhoucke Vincent O
• Szegedy Christian
• Ioffe Sergey

Classifications

• G06F18/24

  Classification techniques · CPC title

• G06V10/454 Primary

  Integrating the filters into a hierarchical structure, e.g. convolutional neural networks [CNN] · CPC title

• G06N3/048

  Activation functions · CPC title

• G06N3/045 Primary

  Combinations of networks · CPC title

• G06N3/044

  Recurrent networks, e.g. Hopfield networks · CPC title