Class MatrixMultiplyOp

Provides outgoing messages for MatrixMultiply(Double[,], Double[,]), given random arguments to the function.

Inheritance

Object

MatrixMultiplyOp

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), "MatrixMultiply", new Type[]{})]
[Quality(QualityBand.Stable)]
public static class MatrixMultiplyOp

Methods

AAverageConditional(DistributionStructArray2D<Gaussian, Double>, DistributionStructArray2D<Gaussian, Double>, DistributionStructArray2D<Gaussian, Double>)

EP message to A.

Declaration

[NotSupported("A MatrixMultiply factor between two Gaussian arrays is not yet implemented for Expectation Propagation.  Try using Variational Message Passing.")]
public static DistributionStructArray2D<Gaussian, double> AAverageConditional(DistributionStructArray2D<Gaussian, double> matrixMultiply, DistributionStructArray2D<Gaussian, double> B, DistributionStructArray2D<Gaussian, double> result)

Parameters

Type	Name	Description
DistributionStructArray2D<Gaussian, Double>	matrixMultiply	Incoming message from matrixMultiply.
DistributionStructArray2D<Gaussian, Double>	B	Incoming message from B. Must be a proper distribution. If any element is uniform, the result will be uniform.
DistributionStructArray2D<Gaussian, Double>	result	Modified to contain the outgoing message.

Returns

Type	Description
DistributionStructArray2D<Gaussian, Double>	result

Remarks

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

Exceptions

Type	Condition
ImproperMessageException	B is not a proper distribution.

AAverageConditional(DistributionStructArray2D<Gaussian, Double>, DistributionStructArray2D<Gaussian, Double>, Double[,], DistributionStructArray2D<Gaussian, Double>)

EP message to A.

Declaration

public static DistributionStructArray2D<Gaussian, double> AAverageConditional(DistributionStructArray2D<Gaussian, double> matrixMultiply, DistributionStructArray2D<Gaussian, double> A, double[, ] B, DistributionStructArray2D<Gaussian, double> result)

Parameters

Type	Name	Description
DistributionStructArray2D<Gaussian, Double>	matrixMultiply	Incoming message from matrixMultiply. Must be a proper distribution. If any element is uniform, the result will be uniform.
DistributionStructArray2D<Gaussian, Double>	A	Incoming message from A. Must be a proper distribution. If any element is uniform, the result will be uniform.
Double[,]	B	Constant value for B.
DistributionStructArray2D<Gaussian, Double>	result	Modified to contain the outgoing message.

Returns

Type	Description
DistributionStructArray2D<Gaussian, Double>	result

Remarks

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

Exceptions

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

AAverageConditional(Double[,], Double[,], DistributionStructArray2D<Gaussian, Double>)

EP message to A.

Declaration

[NotSupported("A MatrixMultiply factor with fixed output is not yet implemented for Expectation Propagation.")]
public static DistributionStructArray2D<Gaussian, double> AAverageConditional(double[, ] matrixMultiply, double[, ] B, DistributionStructArray2D<Gaussian, double> result)

Parameters

Type	Name	Description
Double[,]	matrixMultiply	Constant value for matrixMultiply.
Double[,]	B	Constant value for B.
DistributionStructArray2D<Gaussian, Double>	result	Modified to contain the outgoing message.

Returns

Type	Description
DistributionStructArray2D<Gaussian, Double>	result

Remarks

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

AAverageLogarithm(DistributionStructArray2D<Gaussian, Double>, DistributionStructArray2D<Gaussian, Double>, DistributionStructArray2D<Gaussian, Double>, DistributionStructArray2D<Gaussian, Double>)

VMP message to A.

Declaration

public static DistributionStructArray2D<Gaussian, double> AAverageLogarithm(DistributionStructArray2D<Gaussian, double> matrixMultiply, DistributionStructArray2D<Gaussian, double> A, DistributionStructArray2D<Gaussian, double> B, DistributionStructArray2D<Gaussian, double> to_A)

Parameters

Type	Name	Description
DistributionStructArray2D<Gaussian, Double>	matrixMultiply	Incoming message from matrixMultiply. Must be a proper distribution. If any element is uniform, the result will be uniform.
DistributionStructArray2D<Gaussian, Double>	A	Incoming message from A.
DistributionStructArray2D<Gaussian, Double>	B	Incoming message from B. Must be a proper distribution. If any element is uniform, the result will be uniform.
DistributionStructArray2D<Gaussian, Double>	to_A	Previous outgoing message to A.

Returns

Type	Description
DistributionStructArray2D<Gaussian, Double>	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, matrixMultiply is integrated out before taking the logarithm. The formula is exp(sum_(B) p(B) log(sum_matrixMultiply p(matrixMultiply) factor(matrixMultiply,A,B))).

Exceptions

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

AAverageLogarithm(DistributionStructArray2D<Gaussian, Double>, DistributionStructArray2D<Gaussian, Double>, Double[,], DistributionStructArray2D<Gaussian, Double>)

VMP message to A.

Declaration

public static DistributionStructArray2D<Gaussian, double> AAverageLogarithm(DistributionStructArray2D<Gaussian, double> matrixMultiply, DistributionStructArray2D<Gaussian, double> A, double[, ] B, DistributionStructArray2D<Gaussian, double> to_A)

Parameters

Type	Name	Description
DistributionStructArray2D<Gaussian, Double>	matrixMultiply	Incoming message from matrixMultiply. Must be a proper distribution. If any element is uniform, the result will be uniform.
DistributionStructArray2D<Gaussian, Double>	A	Incoming message from A. Must be a proper distribution. If any element is uniform, the result will be uniform.
Double[,]	B	Constant value for B.
DistributionStructArray2D<Gaussian, Double>	to_A	Previous outgoing message to A.

Returns

Type	Description
DistributionStructArray2D<Gaussian, Double>	The outgoing VMP message to the A argument.

Remarks

The outgoing message is the factor viewed as a function of A with matrixMultiply integrated out. The formula is sum_matrixMultiply p(matrixMultiply) factor(matrixMultiply,A,B).

Exceptions

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

AAverageLogarithm(Double[,])

VMP message to A.

Declaration

[NotSupported("Variational Message Passing does not support a MatrixMultiply factor with fixed output.")]
public static DistributionStructArray2D<Gaussian, double> AAverageLogarithm(double[, ] matrixMultiply)

Parameters

Type	Name	Description
Double[,]	matrixMultiply	Constant value for matrixMultiply.

Returns

Type	Description
DistributionStructArray2D<Gaussian, Double>	The outgoing VMP message to the A argument.

Remarks

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

AverageLogFactor()

Evidence message for VMP.

Declaration

public static double AverageLogFactor()

Returns

Type	Description
Double	Zero.

Remarks

The formula for the result is log(factor(matrixMultiply,A,B)). Adding up these values across all factors and variables gives the log-evidence estimate for VMP.

AverageLogFactor(Double[,], Double[,], Double[,])

Evidence message for VMP.

Declaration

public static double AverageLogFactor(double[, ] matrixMultiply, double[, ] A, double[, ] B)

Parameters

Type	Name	Description
Double[,]	matrixMultiply	Constant value for matrixMultiply.
Double[,]	A	Constant value for A.
Double[,]	B	Constant value for B.

Returns

Type	Description
Double	Zero.

Remarks

The formula for the result is log(factor(matrixMultiply,A,B)). Adding up these values across all factors and variables gives the log-evidence estimate for VMP.

BAverageConditional(DistributionStructArray2D<Gaussian, Double>, DistributionStructArray2D<Gaussian, Double>, DistributionStructArray2D<Gaussian, Double>)

EP message to B.

Declaration

[NotSupported("A MatrixMultiply factor between two Gaussian arrays is not yet implemented for Expectation Propagation.  Try using Variational Message Passing.")]
public static DistributionStructArray2D<Gaussian, double> BAverageConditional(DistributionStructArray2D<Gaussian, double> matrixMultiply, DistributionStructArray2D<Gaussian, double> A, DistributionStructArray2D<Gaussian, double> result)

Parameters

Type	Name	Description
DistributionStructArray2D<Gaussian, Double>	matrixMultiply	Incoming message from matrixMultiply.
DistributionStructArray2D<Gaussian, Double>	A	Incoming message from A. Must be a proper distribution. If any element is uniform, the result will be uniform.
DistributionStructArray2D<Gaussian, Double>	result	Modified to contain the outgoing message.

Returns

Type	Description
DistributionStructArray2D<Gaussian, Double>	result

Remarks

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

Exceptions

Type	Condition
ImproperMessageException	A is not a proper distribution.

BAverageConditional(DistributionStructArray2D<Gaussian, Double>, Double[,], DistributionStructArray2D<Gaussian, Double>, DistributionStructArray2D<Gaussian, Double>)

EP message to B.

Declaration

public static DistributionStructArray2D<Gaussian, double> BAverageConditional(DistributionStructArray2D<Gaussian, double> matrixMultiply, double[, ] A, DistributionStructArray2D<Gaussian, double> B, DistributionStructArray2D<Gaussian, double> result)

Parameters

Type	Name	Description
DistributionStructArray2D<Gaussian, Double>	matrixMultiply	Incoming message from matrixMultiply. Must be a proper distribution. If any element is uniform, the result will be uniform.
Double[,]	A	Constant value for A.
DistributionStructArray2D<Gaussian, Double>	B	Incoming message from B. Must be a proper distribution. If any element is uniform, the result will be uniform.
DistributionStructArray2D<Gaussian, Double>	result	Modified to contain the outgoing message.

Returns

Type	Description
DistributionStructArray2D<Gaussian, Double>	result

Remarks

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

Exceptions

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

BAverageConditional(Double[,], Double[,], DistributionStructArray2D<Gaussian, Double>)

EP message to B.

Declaration

[NotSupported("A MatrixMultiply factor with fixed output is not yet implemented for Expectation Propagation.")]
public static DistributionStructArray2D<Gaussian, double> BAverageConditional(double[, ] matrixMultiply, double[, ] A, DistributionStructArray2D<Gaussian, double> result)

Parameters

Type	Name	Description
Double[,]	matrixMultiply	Constant value for matrixMultiply.
Double[,]	A	Constant value for A.
DistributionStructArray2D<Gaussian, Double>	result	Modified to contain the outgoing message.

Returns

Type	Description
DistributionStructArray2D<Gaussian, Double>	result

Remarks

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

BAverageLogarithm(DistributionStructArray2D<Gaussian, Double>, DistributionStructArray2D<Gaussian, Double>, DistributionStructArray2D<Gaussian, Double>, DistributionStructArray2D<Gaussian, Double>)

VMP message to B.

Declaration

public static DistributionStructArray2D<Gaussian, double> BAverageLogarithm(DistributionStructArray2D<Gaussian, double> matrixMultiply, DistributionStructArray2D<Gaussian, double> A, DistributionStructArray2D<Gaussian, double> B, DistributionStructArray2D<Gaussian, double> to_B)

Parameters

Type	Name	Description
DistributionStructArray2D<Gaussian, Double>	matrixMultiply	Incoming message from matrixMultiply. Must be a proper distribution. If any element is uniform, the result will be uniform.
DistributionStructArray2D<Gaussian, Double>	A	Incoming message from A. Must be a proper distribution. If any element is uniform, the result will be uniform.
DistributionStructArray2D<Gaussian, Double>	B	Incoming message from B. Must be a proper distribution. If any element is uniform, the result will be uniform.
DistributionStructArray2D<Gaussian, Double>	to_B	Previous outgoing message to B.

Returns

Type	Description
DistributionStructArray2D<Gaussian, Double>	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. Because the factor is deterministic, matrixMultiply is integrated out before taking the logarithm. The formula is exp(sum_(A) p(A) log(sum_matrixMultiply p(matrixMultiply) factor(matrixMultiply,A,B))).

Exceptions

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

BAverageLogarithm(DistributionStructArray2D<Gaussian, Double>, Double[,], DistributionStructArray2D<Gaussian, Double>, DistributionStructArray2D<Gaussian, Double>)

VMP message to B.

Declaration

public static DistributionStructArray2D<Gaussian, double> BAverageLogarithm(DistributionStructArray2D<Gaussian, double> matrixMultiply, double[, ] A, DistributionStructArray2D<Gaussian, double> B, DistributionStructArray2D<Gaussian, double> to_B)

Parameters

Type	Name	Description
DistributionStructArray2D<Gaussian, Double>	matrixMultiply	Incoming message from matrixMultiply. Must be a proper distribution. If any element is uniform, the result will be uniform.
Double[,]	A	Constant value for A.
DistributionStructArray2D<Gaussian, Double>	B	Incoming message from B. Must be a proper distribution. If any element is uniform, the result will be uniform.
DistributionStructArray2D<Gaussian, Double>	to_B	Previous outgoing message to B.

Returns

Type	Description
DistributionStructArray2D<Gaussian, Double>	The outgoing VMP message to the B argument.

Remarks

The outgoing message is the factor viewed as a function of B with matrixMultiply integrated out. The formula is sum_matrixMultiply p(matrixMultiply) factor(matrixMultiply,A,B).

Exceptions

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

BAverageLogarithm(Double[,])

VMP message to B.

Declaration

[NotSupported("Variational Message Passing does not support a MatrixMultiply factor with fixed output.")]
public static DistributionStructArray2D<Gaussian, double> BAverageLogarithm(double[, ] matrixMultiply)

Parameters

Type	Name	Description
Double[,]	matrixMultiply	Constant value for matrixMultiply.

Returns

Type	Description
DistributionStructArray2D<Gaussian, Double>	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.

LogAverageFactor(DistributionStructArray2D<Gaussian, Double>, DistributionStructArray2D<Gaussian, Double>, Double[,])

Evidence message for EP.

Declaration

public static double LogAverageFactor(DistributionStructArray2D<Gaussian, double> matrixMultiply, DistributionStructArray2D<Gaussian, double> A, double[, ] B)

Parameters

Type	Name	Description
DistributionStructArray2D<Gaussian, Double>	matrixMultiply	Incoming message from matrixMultiply.
DistributionStructArray2D<Gaussian, Double>	A	Incoming message from A.
Double[,]	B	Constant value for B.

Returns

Type	Description
Double	Logarithm of the factor's average value across the given argument distributions.

Remarks

The formula for the result is log(sum_(matrixMultiply,A) p(matrixMultiply,A) factor(matrixMultiply,A,B)).

LogAverageFactor(DistributionStructArray2D<Gaussian, Double>, Double[,], DistributionStructArray2D<Gaussian, Double>)

Evidence message for EP.

Declaration

public static double LogAverageFactor(DistributionStructArray2D<Gaussian, double> matrixMultiply, double[, ] A, DistributionStructArray2D<Gaussian, double> B)

Parameters

Type	Name	Description
DistributionStructArray2D<Gaussian, Double>	matrixMultiply	Incoming message from matrixMultiply.
Double[,]	A	Constant value for A.
DistributionStructArray2D<Gaussian, Double>	B	Incoming message from B.

Returns

Type	Description
Double	Logarithm of the factor's average value across the given argument distributions.

Remarks

The formula for the result is log(sum_(matrixMultiply,B) p(matrixMultiply,B) factor(matrixMultiply,A,B)).

LogAverageFactor(Double[,], DistributionStructArray2D<Gaussian, Double>, Double[,])

Evidence message for EP.

Declaration

public static double LogAverageFactor(double[, ] matrixMultiply, DistributionStructArray2D<Gaussian, double> A, double[, ] B)

Parameters

Type	Name	Description
Double[,]	matrixMultiply	Constant value for matrixMultiply.
DistributionStructArray2D<Gaussian, Double>	A	Incoming message from A.
Double[,]	B	Constant value for B.

Returns

Type	Description
Double	Logarithm of the factor's average value across the given argument distributions.

Remarks

The formula for the result is log(sum_(A) p(A) factor(matrixMultiply,A,B)).

LogAverageFactor(Double[,], Double[,], DistributionStructArray2D<Gaussian, Double>)

Evidence message for EP.

Declaration

public static double LogAverageFactor(double[, ] matrixMultiply, double[, ] A, DistributionStructArray2D<Gaussian, double> B)

Parameters

Type	Name	Description
Double[,]	matrixMultiply	Constant value for matrixMultiply.
Double[,]	A	Constant value for A.
DistributionStructArray2D<Gaussian, Double>	B	Incoming message from B.

Returns

Type	Description
Double	Logarithm of the factor's average value across the given argument distributions.

Remarks

The formula for the result is log(sum_(B) p(B) factor(matrixMultiply,A,B)).

LogAverageFactor(Double[,], Double[,], Double[,])

Evidence message for EP.

Declaration

public static double LogAverageFactor(double[, ] matrixMultiply, double[, ] A, double[, ] B)

Parameters

Type	Name	Description
Double[,]	matrixMultiply	Constant value for matrixMultiply.
Double[,]	A	Constant value for A.
Double[,]	B	Constant value for B.

Returns

Type	Description
Double	Logarithm of the factor's average value across the given argument distributions.

Remarks

The formula for the result is log(factor(matrixMultiply,A,B)).

LogEvidenceRatio(DistributionStructArray2D<Gaussian, Double>, DistributionStructArray2D<Gaussian, Double>, Double[,])

Evidence message for EP.

Declaration

public static double LogEvidenceRatio(DistributionStructArray2D<Gaussian, double> matrixMultiply, DistributionStructArray2D<Gaussian, double> A, double[, ] B)

Parameters

Type	Name	Description
DistributionStructArray2D<Gaussian, Double>	matrixMultiply	Incoming message from matrixMultiply.
DistributionStructArray2D<Gaussian, Double>	A	Incoming message from A.
Double[,]	B	Constant value for B.

Returns

Type	Description
Double	Logarithm of the factor's contribution the EP model evidence.

Remarks

The formula for the result is log(sum_(matrixMultiply,A) p(matrixMultiply,A) factor(matrixMultiply,A,B) / sum_matrixMultiply p(matrixMultiply) messageTo(matrixMultiply)). Adding up these values across all factors and variables gives the log-evidence estimate for EP.

LogEvidenceRatio(DistributionStructArray2D<Gaussian, Double>, Double[,], DistributionStructArray2D<Gaussian, Double>)

Evidence message for EP.

Declaration

public static double LogEvidenceRatio(DistributionStructArray2D<Gaussian, double> matrixMultiply, double[, ] A, DistributionStructArray2D<Gaussian, double> B)

Parameters

Type	Name	Description
DistributionStructArray2D<Gaussian, Double>	matrixMultiply	Incoming message from matrixMultiply.
Double[,]	A	Constant value for A.
DistributionStructArray2D<Gaussian, Double>	B	Incoming message from B.

Returns

Type	Description
Double	Logarithm of the factor's contribution the EP model evidence.

Remarks

The formula for the result is log(sum_(matrixMultiply,B) p(matrixMultiply,B) factor(matrixMultiply,A,B) / sum_matrixMultiply p(matrixMultiply) messageTo(matrixMultiply)). Adding up these values across all factors and variables gives the log-evidence estimate for EP.

LogEvidenceRatio(Double[,], DistributionStructArray2D<Gaussian, Double>, Double[,])

Evidence message for EP.

Declaration

public static double LogEvidenceRatio(double[, ] matrixMultiply, DistributionStructArray2D<Gaussian, double> A, double[, ] B)

Parameters

Type	Name	Description
Double[,]	matrixMultiply	Constant value for matrixMultiply.
DistributionStructArray2D<Gaussian, Double>	A	Incoming message from A.
Double[,]	B	Constant value for B.

Returns

Type	Description
Double	Logarithm of the factor's contribution the EP model evidence.

Remarks

The formula for the result is log(sum_(A) p(A) factor(matrixMultiply,A,B)). Adding up these values across all factors and variables gives the log-evidence estimate for EP.

LogEvidenceRatio(Double[,], Double[,], DistributionStructArray2D<Gaussian, Double>)

Evidence message for EP.

Declaration

public static double LogEvidenceRatio(double[, ] matrixMultiply, double[, ] A, DistributionStructArray2D<Gaussian, double> B)

Parameters

Type	Name	Description
Double[,]	matrixMultiply	Constant value for matrixMultiply.
Double[,]	A	Constant value for A.
DistributionStructArray2D<Gaussian, Double>	B	Incoming message from B.

Returns

Type	Description
Double	Logarithm of the factor's contribution the EP model evidence.

Remarks

The formula for the result is log(sum_(B) p(B) factor(matrixMultiply,A,B)). Adding up these values across all factors and variables gives the log-evidence estimate for EP.

LogEvidenceRatio(Double[,], Double[,], Double[,])

Evidence message for EP.

Declaration

public static double LogEvidenceRatio(double[, ] matrixMultiply, double[, ] A, double[, ] B)

Parameters

Type	Name	Description
Double[,]	matrixMultiply	Constant value for matrixMultiply.
Double[,]	A	Constant value for A.
Double[,]	B	Constant value for B.

Returns

Type	Description
Double	Logarithm of the factor's contribution the EP model evidence.

Remarks

The formula for the result is log(factor(matrixMultiply,A,B)). Adding up these values across all factors and variables gives the log-evidence estimate for EP.

MatrixMultiplyAverageConditional(DistributionStructArray2D<Gaussian, Double>, DistributionStructArray2D<Gaussian, Double>, DistributionStructArray2D<Gaussian, Double>)

EP message to matrixMultiply.

Declaration

[NotSupported("A MatrixMultiply factor between two Gaussian arrays is not yet implemented for Expectation Propagation.  Try using Variational Message Passing.")]
public static DistributionStructArray2D<Gaussian, double> MatrixMultiplyAverageConditional(DistributionStructArray2D<Gaussian, double> A, DistributionStructArray2D<Gaussian, double> B, DistributionStructArray2D<Gaussian, double> result)

Parameters

Type	Name	Description
DistributionStructArray2D<Gaussian, Double>	A	Incoming message from A. Must be a proper distribution. If any element is uniform, the result will be uniform.
DistributionStructArray2D<Gaussian, Double>	B	Incoming message from B. Must be a proper distribution. If any element is uniform, the result will be uniform.
DistributionStructArray2D<Gaussian, Double>	result	Modified to contain the outgoing message.

Returns

Type	Description
DistributionStructArray2D<Gaussian, Double>	result

Remarks

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

Exceptions

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

MatrixMultiplyAverageConditional(DistributionStructArray2D<Gaussian, Double>, Double[,], DistributionStructArray2D<Gaussian, Double>)

EP message to matrixMultiply.

Declaration

public static DistributionStructArray2D<Gaussian, double> MatrixMultiplyAverageConditional(DistributionStructArray2D<Gaussian, double> A, double[, ] B, DistributionStructArray2D<Gaussian, double> result)

Parameters

Type	Name	Description
DistributionStructArray2D<Gaussian, Double>	A	Incoming message from A. Must be a proper distribution. If any element is uniform, the result will be uniform.
Double[,]	B	Constant value for B.
DistributionStructArray2D<Gaussian, Double>	result	Modified to contain the outgoing message.

Returns

Type	Description
DistributionStructArray2D<Gaussian, Double>	result

Remarks

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

Exceptions

Type	Condition
ImproperMessageException	A is not a proper distribution.

MatrixMultiplyAverageConditional(Double[,], DistributionStructArray2D<Gaussian, Double>, DistributionStructArray2D<Gaussian, Double>)

EP message to matrixMultiply.

Declaration

public static DistributionStructArray2D<Gaussian, double> MatrixMultiplyAverageConditional(double[, ] A, DistributionStructArray2D<Gaussian, double> B, DistributionStructArray2D<Gaussian, double> result)

Parameters

Type	Name	Description
Double[,]	A	Constant value for A.
DistributionStructArray2D<Gaussian, Double>	B	Incoming message from B. Must be a proper distribution. If any element is uniform, the result will be uniform.
DistributionStructArray2D<Gaussian, Double>	result	Modified to contain the outgoing message.

Returns

Type	Description
DistributionStructArray2D<Gaussian, Double>	result

Remarks

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

Exceptions

Type	Condition
ImproperMessageException	B is not a proper distribution.

MatrixMultiplyAverageLogarithm(DistributionArray2D<Gaussian>, DistributionArray2D<Gaussian>, DistributionStructArray2D<Gaussian, Double>)

VMP message to matrixMultiply.

Declaration

public static DistributionStructArray2D<Gaussian, double> MatrixMultiplyAverageLogarithm(DistributionArray2D<Gaussian> A, DistributionArray2D<Gaussian> B, DistributionStructArray2D<Gaussian, double> result)

Parameters

Type	Name	Description
DistributionArray2D<Gaussian>	A	Incoming message from A. Must be a proper distribution. If any element is uniform, the result will be uniform.
DistributionArray2D<Gaussian>	B	Incoming message from B. Must be a proper distribution. If any element is uniform, the result will be uniform.
DistributionStructArray2D<Gaussian, Double>	result	Modified to contain the outgoing message.

Returns

Type	Description
DistributionStructArray2D<Gaussian, Double>	result

Remarks

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

Exceptions

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

MatrixMultiplyAverageLogarithm(DistributionStructArray2D<Gaussian, Double>, Double[,], DistributionStructArray2D<Gaussian, Double>)

VMP message to matrixMultiply.

Declaration

public static DistributionStructArray2D<Gaussian, double> MatrixMultiplyAverageLogarithm(DistributionStructArray2D<Gaussian, double> A, double[, ] B, DistributionStructArray2D<Gaussian, double> result)

Parameters

Type	Name	Description
DistributionStructArray2D<Gaussian, Double>	A	Incoming message from A. Must be a proper distribution. If any element is uniform, the result will be uniform.
Double[,]	B	Constant value for B.
DistributionStructArray2D<Gaussian, Double>	result	Modified to contain the outgoing message.

Returns

Type	Description
DistributionStructArray2D<Gaussian, Double>	result

Remarks

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

Exceptions

Type	Condition
ImproperMessageException	A is not a proper distribution.

MatrixMultiplyAverageLogarithm(Double[,], DistributionStructArray2D<Gaussian, Double>, DistributionStructArray2D<Gaussian, Double>)

VMP message to matrixMultiply.

Declaration

public static DistributionStructArray2D<Gaussian, double> MatrixMultiplyAverageLogarithm(double[, ] A, DistributionStructArray2D<Gaussian, double> B, DistributionStructArray2D<Gaussian, double> result)

Parameters

Type	Name	Description
Double[,]	A	Constant value for A.
DistributionStructArray2D<Gaussian, Double>	B	Incoming message from B. Must be a proper distribution. If any element is uniform, the result will be uniform.
DistributionStructArray2D<Gaussian, Double>	result	Modified to contain the outgoing message.

Returns

Type	Description
DistributionStructArray2D<Gaussian, Double>	result

Remarks

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

Exceptions

Type	Condition
ImproperMessageException	B is not a proper distribution.

MatrixMultiplyAverageLogarithmInit(DistributionArray2D<Gaussian>, DistributionArray2D<Gaussian>)

Declaration

public static DistributionStructArray2D<Gaussian, double> MatrixMultiplyAverageLogarithmInit(DistributionArray2D<Gaussian> A, DistributionArray2D<Gaussian> B)

Parameters

Type	Name	Description
DistributionArray2D<Gaussian>	A	Incoming message from A.
DistributionArray2D<Gaussian>	B	Incoming message from B.

Returns

Type	Description
DistributionStructArray2D<Gaussian, Double>

Remarks

MatrixMultiplyAverageLogarithmInit(DistributionArray2D<Gaussian>, Double[,])

Declaration

public static DistributionStructArray2D<Gaussian, double> MatrixMultiplyAverageLogarithmInit(DistributionArray2D<Gaussian> A, double[, ] B)

Parameters

Type	Name	Description
DistributionArray2D<Gaussian>	A	Incoming message from A.
Double[,]	B	Constant value for B.

Returns

Type	Description
DistributionStructArray2D<Gaussian, Double>

Remarks

MatrixMultiplyAverageLogarithmInit(Double[,], DistributionArray2D<Gaussian>)

Declaration

public static DistributionStructArray2D<Gaussian, double> MatrixMultiplyAverageLogarithmInit(double[, ] A, DistributionArray2D<Gaussian> B)

Parameters

Type	Name	Description
Double[,]	A	Constant value for A.
DistributionArray2D<Gaussian>	B	Incoming message from B.

Returns

Type	Description
DistributionStructArray2D<Gaussian, Double>

Remarks