Class BipolarSigmoid

Namespace

NeuralNetworks.Operations.ActivationFunctions

Assembly

NeuralNetworks.dll

Represents a bipolar sigmoid activation operation that applies a scaled sigmoid function shifted to the range [-0.5, 0.5].

public class BipolarSigmoid : ActivationFunction2D

Inheritance

object

Operation

Operation<float[,], float[,]>

Operation2D

ActivationFunction2D

BipolarSigmoid

Inherited Members

Operation2D.EnsureSameShapeForInput(float[,], float[,])

Operation2D.EnsureSameShapeForOutput(float[,], float[,])

Operation<float[,], float[,]>.Input

Operation<float[,], float[,]>.Output

Operation<float[,], float[,]>.Forward(float[,], bool)

Operation<float[,], float[,]>.Backward(float[,])

Operation<float[,], float[,]>.EnsureSameShapeForInput(float[,], float[,])

Operation<float[,], float[,]>.EnsureSameShapeForOutput(float[,], float[,])

Operation<float[,], float[,]>.CalcOutput(bool)

Operation<float[,], float[,]>.CalcInputGradient(float[,])

Operation<float[,], float[,]>.Forward(object, bool)

Operation<float[,], float[,]>.Backward(object)

Operation<float[,], float[,]>.GetOutputType()

Operation<float[,], float[,]>.GetInputType()

Operation<float[,], float[,]>.CloneBase()

Operation<float[,], float[,]>.Clone()

Operation.SetRegistered()

Operation.GetOutputType()

Operation.GetInputType()

Operation.Forward(object, bool)

Operation.Backward(object)

object.Equals(object)

object.Equals(object, object)

object.GetHashCode()

object.GetType()

object.MemberwiseClone()

object.ReferenceEquals(object, object)

Remarks

The bipolar sigmoid function is commonly used in neural networks to introduce non-linearity while centering the output around zero. This operation computes output as scale × (σ(x) – 0.5), where σ(x) is the standard sigmoid function. Setting an appropriate scale can affect the gradient flow and the range of activations.

With scale = 2: y = 2σ(x) − 1, which is exactly the bipolar (a.k.a. symmetric/zero-centered) sigmoid. It’s mathematically identical to tanh(x/2). For general scale s: y = (s/2) · tanh(x/2), i.e., a scaled tanh.Range is [−s/2, s/2].

Constructors

BipolarSigmoid(float)

Represents a bipolar sigmoid activation operation that applies a scaled sigmoid function shifted to the range [-0.5, 0.5].

public BipolarSigmoid(float scale)

Parameters

scale float

The scaling factor applied to the output of the sigmoid function. Must be non-zero.

Remarks

The bipolar sigmoid function is commonly used in neural networks to introduce non-linearity while centering the output around zero. This operation computes output as scale × (σ(x) – 0.5), where σ(x) is the standard sigmoid function. Setting an appropriate scale can affect the gradient flow and the range of activations.

With scale = 2: y = 2σ(x) − 1, which is exactly the bipolar (a.k.a. symmetric/zero-centered) sigmoid. It’s mathematically identical to tanh(x/2). For general scale s: y = (s/2) · tanh(x/2), i.e., a scaled tanh.Range is [−s/2, s/2].

Methods

CalcInputGradient(float[,])

Calculates input gradient.

protected override float[,] CalcInputGradient(float[,] outputGradient)

Parameters

outputGradient float[,]

Returns

float[,]

Remarks

Based on outputGradient, calculates changes in input.

CalcOutput(bool)

Computes output.

protected override float[,] CalcOutput(bool inference)

Parameters

inference bool

Returns

float[,]

ToString()

Returns a string that represents the current object.

public override string ToString()

Returns

string

A string that represents the current object.