moefactory

moefactory

Construct a mixture of experts distribution structure

Syntax

D = mixturefactory(GateD, ExpertD)

Description

D = mixturefactory(GateD, ExpertD) returns a structure representing a mixture of experts distribution. GatetD is a gate structure. ExpertD defines the expert distributions and it may be either a conditional distribution structure or a cell array of conditional distribution structures on the same data space.

When ExpertD is a conditional distribution structure, it defines the expert component distribution type. The number of experts equals GateD.num().

When ExpertD is a cell array of conditional distribution structures, the function constructs a heterogeneous mixture of experts distribution where each expert may be of a different distribution type. The number of elements in the cell array should equal GateD.num().

Distribution Parameters

  • D (num-by-1 cell array of distribution parameter structures) : Contains the parameters for each expert.
  • G : Sturucture containing the parameters of the gate.

Probability Density Function

The distribution has the following density:

$$ f(y | x) =  \sum_{k=1}^{num}{ \pi_k(x) g_k(y|x)} $$

where $\pi_k(x)$ is the k-th gate component defined by the gate distribution and $g_k(y|x)$ is the k-th expert component.

name

See distribution structure common members.

M

See distribution structure common members.

expert

expert distributions

Syntax

D_k = D.expert(idx)

Description

D_k = D.expert(idx) where idx is the index of a component, gives the component distribution structure at idx.

D_k = D.expert(idx) where idx is an index vector with more than one element, returns a cell array of the distribution structures indexed in idx.

Valid range for the indices is from 1 to D.numtotal().

gate

Gate distribution

num

Number of components (excluding any fixed components)

Syntax

num = D.num()

dim

See distribution structure common members.

datadim

See distribution structure common members.

llvec

See distribution structure common members.

ll

See distribution structure common members.

llgrad

See distribution structure common members.

llgraddata

See distribution structure common members.

predict

pdf

See distribution structure common members.

sample

See distribution structure common members.

randparam

See distribution structure common members.

init

See distribution structure common members.

estimatedefault

penalizerparam

See distribution structure common members.

penalizercost

See distribution structure common members.

penalizergrad

See distribution structure common members.

sumparam

See distribution structure common members.

scaleparam

See distribution structure common members.

sumgrad

See distribution structure common members.

scalegrad

See distribution structure common members.

entropy

See distribution structure common members.

kl

See distribution structure common members.

AICc

See distribution structure common members.

BIC

See distribution structure common members.

display

See distribution structure common members.