crossChannelNormalizationLayer

Channel-wise local response normalization layer

Description

A channel-wise local response (cross-channel) normalization layer carries out channel-wise normalization.

Creation

Syntax

layer = crossChannelNormalizationLayer(windowChannelSize)

layer = crossChannelNormalizationLayer(windowChannelSize,Name,Value)

Description

layer = crossChannelNormalizationLayer(windowChannelSize) creates a channel-wise local response normalization layer and sets the WindowChannelSize property.

layer = crossChannelNormalizationLayer(windowChannelSize,Name,Value) sets the optional properties WindowChannelSize, Alpha, Beta, K, and Name using name-value pairs. For example, crossChannelNormalizationLayer(5,'K',1) creates a local response normalization layer for channel-wise normalization with a window size of 5 and K hyperparameter 1. You can specify multiple name-value pairs. Enclose each property name in single quotes.

Properties

Cross-Channel Normalization

WindowChannelSize — Size of the channel window

positive integer less than or equal to 16

Size of the channel window, which controls the number of channels that are used for the normalization of each element, specified as a positive integer less than or equal to 16.

If WindowChannelSize is even, then the window is asymmetric. The software looks at the previous floor((w-1)/2) channels and the following floor(w/2) channels. For example, if WindowChannelSize is 4, then the layer normalizes each element by its neighbor in the previous channel and by its neighbors in the next two channels.

Example: 5

Alpha — α hyperparameter in normalization

0.0001 (default) | numeric scalar

α hyperparameter in the normalization (the multiplier term), specified as a numeric scalar.

Example: 0.0002

Beta — β hyperparameter in normalization

0.75 (default) | numeric scalar

β hyperparameter in the normalization, specified as a numeric scalar. The value of Beta must be greater than or equal to 0.01.

Example: 0.8

K — K hyperparameter in the normalization

2 (default) | numeric scalar

K hyperparameter in the normalization, specified as a numeric scalar. The value of K must be greater than or equal to 10^-5.

Example: 2.5

Layer

Name — Layer name

'' (default) | character vector | string scalar

Layer name, specified as a character vector or a string scalar. For Layer array input, the trainnet and dlnetwork functions automatically assign names to unnamed layers.

The CrossChannelNormalizationLayer object stores this property as a character vector.

Data Types: char | string

NumInputs — Number of inputs

Read-only: 1 (default)

This property is read-only.

Number of inputs to the layer, stored as 1. This layer accepts a single input only.

Data Types: double

InputNames — Input names

Read-only: {'in'} (default)

This property is read-only.

Input names, stored as {'in'}. This layer accepts a single input only.

Data Types: cell

NumOutputs — Number of outputs

Read-only: 1 (default)

This property is read-only.

Number of outputs from the layer, stored as 1. This layer has a single output only.

Data Types: double

OutputNames — Output names

Read-only: {'out'} (default)

This property is read-only.

Output names, stored as {'out'}. This layer has a single output only.

Data Types: cell

Examples

Create Local Response Normalization Layer

Create a local response normalization layer for channel-wise normalization, where a window of five channels normalizes each element, and the additive constant for the normalizer $\mathit{K}$ is 1.

layer = crossChannelNormalizationLayer(5,K=1)

layer = 
  CrossChannelNormalizationLayer with properties:

                 Name: ''

   Hyperparameters
    WindowChannelSize: 5
                Alpha: 1.0000e-04
                 Beta: 0.7500
                    K: 1

Include a local response normalization layer in a Layer array.

layers = [ ...
    imageInputLayer([28 28 1])
    convolution2dLayer(5,20)
    reluLayer
    crossChannelNormalizationLayer(3)
    fullyConnectedLayer(10)
    softmaxLayer]

layers = 
  6×1 Layer array with layers:

     1   ''   Image Input                   28×28×1 images with 'zerocenter' normalization
     2   ''   2-D Convolution               20 5×5 convolutions with stride [1  1] and padding [0  0  0  0]
     3   ''   ReLU                          ReLU
     4   ''   Cross Channel Normalization   cross channel normalization with 3 channels per element
     5   ''   Fully Connected               10 fully connected layer
     6   ''   Softmax                       softmax

Algorithms

Cross Channel Normalization Layer

A channel-wise local response (cross-channel) normalization layer carries out channel-wise normalization.

This layer performs a channel-wise local response normalization. It usually follows the ReLU activation layer. This layer replaces each element with a normalized value it obtains using the elements from a certain number of neighboring channels (elements in the normalization window). That is, for each element $$x$$ in the input, the trainnet function computes a normalized value $$x^{\text{'}}$$ using

$$x^{\text{'}}=\frac{x}{{\left(K+\frac{\alpha *ss}{windowChannelSize}\right)}^{\beta }},$$

where K, α, and β are the hyperparameters in the normalization, and ss is the sum of squares of the elements in the normalization window [1]. You must specify the size of the normalization window using the windowChannelSize argument of the crossChannelNormalizationLayer function. You can also specify the hyperparameters using the Alpha, Beta, and K name-value pair arguments.

The previous normalization formula is slightly different than what is presented in [1]. You can obtain the equivalent formula by multiplying the alpha value by the windowChannelSize.

Layer Input and Output Formats

Layers in a layer array or layer graph pass data to subsequent layers as formatted dlarray objects. The format of a dlarray object is a string of characters in which each character describes the corresponding dimension of the data. The format consists of one or more of these characters:

• "S" — Spatial

• "C" — Channel

• "B" — Batch

• "T" — Time

• "U" — Unspecified

For example, you can describe 2-D image data that is represented as a 4-D array, where the first two dimensions correspond to the spatial dimensions of the images, the third dimension corresponds to the channels of the images, and the fourth dimension corresponds to the batch dimension, as having the format "SSCB" (spatial, spatial, channel, batch).

You can interact with these dlarray objects in automatic differentiation workflows, such as those for developing a custom layer, using a functionLayer object, or using the forward and predict functions with dlnetwork objects.

This table shows the supported input formats of CrossChannelNormalizationLayer objects and the corresponding output format. If the software passes the output of the layer to a custom layer that does not inherit from the nnet.layer.Formattable class, or a FunctionLayer object with the Formattable property set to 0 (false), then the layer receives an unformatted dlarray object with dimensions ordered according to the formats in this table. The formats listed here are only a subset. The layer may support additional formats such as formats with additional "S" (spatial) or "U" (unspecified) dimensions.

Input Format	Output Format
"SSCB" (spatial, spatial, channel, batch)	"SSCB" (spatial, spatial, channel, batch)
"SSC" (spatial, spatial, channel)	"SSC" (spatial, spatial, channel)
"SSSB" (spatial, spatial, spatial, batch)	"SSSB" (spatial, spatial, spatial, batch)

In dlnetwork objects, CrossChannelNormalizationLayer objects also support these input and output format combinations.

Input Format	Output Format
"SSCBT" (spatial, spatial, channel, batch, time)	"SSCBT" (spatial, spatial, channel, batch, time)
"SSCT" (spatial, spatial, channel, time)	"SSCT" (spatial, spatial, channel, time)
"SSSBT" (spatial, spatial, spatial, batch, time)	"SSSBT" (spatial, spatial, spatial, batch, time)

References

[1] Krizhevsky, A., I. Sutskever, and G. E. Hinton. "ImageNet Classification with Deep Convolutional Neural Networks." Advances in Neural Information Processing Systems. Vol 25, 2012.

Extended Capabilities

C/C++ Code Generation
Generate C and C++ code using MATLAB® Coder™.

GPU Code Generation
Generate CUDA® code for NVIDIA® GPUs using GPU Coder™.

Version History

Introduced in R2016a