Class ProductExpOp

Provides outgoing messages for ProductExp(Double, Double), given random arguments to the function.

Inheritance

Object

ProductExpOp

Inherited Members

Object.Equals(Object)

Object.Equals(Object, Object)

Object.GetHashCode()

Object.GetType()

Object.MemberwiseClone()

Object.ReferenceEquals(Object, Object)

Object.ToString()

Namespace: Microsoft.ML.Probabilistic.Factors

Assembly: Microsoft.ML.Probabilistic.dll

Syntax

[FactorMethod(typeof(Factor), "ProductExp", new Type[]{typeof(double), typeof(double)})]
[Quality(QualityBand.Experimental)]
public static class ProductExpOp

Methods

AAverageLogarithm(Gaussian, Gaussian)

VMP message to a.

Declaration

public static Gaussian AAverageLogarithm(Gaussian ProductExp, Gaussian B)

Parameters

Type	Name	Description
Gaussian	ProductExp	Incoming message from productExp. Must be a proper distribution. If uniform, the result will be uniform.
Gaussian	B	Incoming message from b. Must be a proper distribution. If uniform, the result will be uniform.

Returns

Type	Description
Gaussian	The outgoing VMP message to the a argument.

Remarks

The outgoing message is the exponential of the average log-factor value, where the average is over all arguments except a. Because the factor is deterministic, productExp is integrated out before taking the logarithm. The formula is exp(sum_(b) p(b) log(sum_productExp p(productExp) factor(productExp,a,b))).

Exceptions

Type	Condition
ImproperMessageException	ProductExp is not a proper distribution.
ImproperMessageException	B is not a proper distribution.

AAverageLogarithm(Double, Gaussian)

VMP message to a.

Declaration

[NotSupported("Variational Message Passing does not support a Product factor with fixed output and two random inputs.")]
public static Gaussian AAverageLogarithm(double ProductExp, Gaussian B)

Parameters

Type	Name	Description
Double	ProductExp	Constant value for productExp.
Gaussian	B	Incoming message from b. Must be a proper distribution. If uniform, the result will be uniform.

Returns

Type	Description
Gaussian	The outgoing VMP message to the a argument.

Remarks

The outgoing message is the exponential of the average log-factor value, where the average is over all arguments except a. The formula is exp(sum_(b) p(b) log(factor(productExp,a,b))).

Exceptions

Type	Condition
ImproperMessageException	B is not a proper distribution.

BAverageLogarithm(Gaussian, Gaussian, Gaussian, Gaussian)

VMP message to b.

Declaration

public static Gaussian BAverageLogarithm(Gaussian ProductExp, Gaussian A, Gaussian B, Gaussian result)

Parameters

Type	Name	Description
Gaussian	ProductExp	Incoming message from productExp. Must be a proper distribution. If uniform, the result will be uniform.
Gaussian	A	Incoming message from a. Must be a proper distribution. If uniform, the result will be uniform.
Gaussian	B	Incoming message from b. Must be a proper distribution. If uniform, the result will be uniform.
Gaussian	result	Modified to contain the outgoing message.

Returns

Type	Description
Gaussian	result

Remarks

The outgoing message is the exponential of the average log-factor value, where the average is over all arguments except b. Because the factor is deterministic, productExp is integrated out before taking the logarithm. The formula is exp(sum_(a) p(a) log(sum_productExp p(productExp) factor(productExp,a,b))).

Exceptions

Type	Condition
ImproperMessageException	ProductExp is not a proper distribution.
ImproperMessageException	A is not a proper distribution.
ImproperMessageException	B is not a proper distribution.

BAverageLogarithm(Gaussian, Gaussian, Gaussian, NonconjugateGaussian, NonconjugateGaussian)

VMP message to b.

Declaration

public static NonconjugateGaussian BAverageLogarithm(Gaussian ProductExp, Gaussian A, Gaussian B, NonconjugateGaussian to_B, NonconjugateGaussian result)

Parameters

Type	Name	Description
Gaussian	ProductExp	Incoming message from productExp. Must be a proper distribution. If uniform, the result will be uniform.
Gaussian	A	Incoming message from a. Must be a proper distribution. If uniform, the result will be uniform.
Gaussian	B	Incoming message from b. Must be a proper distribution. If uniform, the result will be uniform.
NonconjugateGaussian	to_B	Previous outgoing message to B.
NonconjugateGaussian	result	Modified to contain the outgoing message.

Returns

Type	Description
NonconjugateGaussian	result

Remarks

The outgoing message is the exponential of the average log-factor value, where the average is over all arguments except b. Because the factor is deterministic, productExp is integrated out before taking the logarithm. The formula is exp(sum_(a) p(a) log(sum_productExp p(productExp) factor(productExp,a,b))).

Exceptions

Type	Condition
ImproperMessageException	ProductExp is not a proper distribution.
ImproperMessageException	A is not a proper distribution.
ImproperMessageException	B is not a proper distribution.

BAverageLogarithm(Gaussian, Double, Gaussian, NonconjugateGaussian, NonconjugateGaussian)

VMP message to b.

Declaration

public static NonconjugateGaussian BAverageLogarithm(Gaussian ProductExp, double A, Gaussian B, NonconjugateGaussian to_B, NonconjugateGaussian result)

Parameters

Type	Name	Description
Gaussian	ProductExp	Incoming message from productExp. Must be a proper distribution. If uniform, the result will be uniform.
Double	A	Constant value for a.
Gaussian	B	Incoming message from b. Must be a proper distribution. If uniform, the result will be uniform.
NonconjugateGaussian	to_B	Previous outgoing message to B.
NonconjugateGaussian	result	Modified to contain the outgoing message.

Returns

Type	Description
NonconjugateGaussian	result

Remarks

The outgoing message is the factor viewed as a function of b with productExp integrated out. The formula is sum_productExp p(productExp) factor(productExp,a,b).

Exceptions

Type	Condition
ImproperMessageException	ProductExp is not a proper distribution.
ImproperMessageException	B is not a proper distribution.

BAverageLogarithm(Double, Gaussian)

VMP message to b.

Declaration

[NotSupported("Variational Message Passing does not support a Product factor with fixed output and two random inputs.")]
public static NonconjugateGaussian BAverageLogarithm(double ProductExp, Gaussian A)

Parameters

Type	Name	Description
Double	ProductExp	Constant value for productExp.
Gaussian	A	Incoming message from a.

Returns

Type	Description
NonconjugateGaussian	The outgoing VMP message to the b argument.

Remarks

The outgoing message is the exponential of the average log-factor value, where the average is over all arguments except b. The formula is exp(sum_(a) p(a) log(factor(productExp,a,b))).

BAverageLogarithm(Double, Double)

VMP message to b.

Declaration

public static Gamma BAverageLogarithm(double ProductExp, double A)

Parameters

Type	Name	Description
Double	ProductExp	Constant value for productExp.
Double	A	Constant value for a.

Returns

Type	Description
Gamma	The outgoing VMP message to the b argument.

Remarks

The outgoing message is the factor viewed as a function of b conditioned on the given values.

ProductExpAverageLogarithm(Gaussian, Gaussian)

VMP message to productExp.

Declaration

public static Gaussian ProductExpAverageLogarithm(Gaussian A, Gaussian B)

Parameters

Type	Name	Description
Gaussian	A	Incoming message from a. Must be a proper distribution. If uniform, the result will be uniform.
Gaussian	B	Incoming message from b. Must be a proper distribution. If uniform, the result will be uniform.

Returns

Type	Description
Gaussian	The outgoing VMP message to the productExp argument.

Remarks

The outgoing message is a distribution matching the moments of productExp as the random arguments are varied. The formula is proj[sum_(a,b) p(a,b) factor(productExp,a,b)].

Exceptions

Type	Condition
ImproperMessageException	A is not a proper distribution.
ImproperMessageException	B is not a proper distribution.

ProductExpAverageLogarithm(Double, Gaussian)

VMP message to productExp.

Declaration

public static Gaussian ProductExpAverageLogarithm(double A, Gaussian B)

Parameters

Type	Name	Description
Double	A	Constant value for a.
Gaussian	B	Incoming message from b. Must be a proper distribution. If uniform, the result will be uniform.

Returns

Type	Description
Gaussian	The outgoing VMP message to the productExp argument.

Remarks

The outgoing message is a distribution matching the moments of productExp as the random arguments are varied. The formula is proj[sum_(b) p(b) factor(productExp,a,b)].

Exceptions

Type	Condition
ImproperMessageException	B is not a proper distribution.