mmtsneP {mmtsne}

Multiple maps t-SNE with symmetric probability matrix

Description

mmtsneP estimates a multiple maps t-distributed stochastic neighbor embedding (multiple maps t-SNE) model.

Usage

mmtsneP(P, no_maps, no_dims = 2, max_iter = 500, momentum = 0.5,
  final_momentum = 0.8, mom_switch_iter = 250, eps = 1e-07)

Arguments

P

An N x N symmetric joint probability distribution matrix. These can be constructed from an N by D matrix with x2p and p2sp. Alternatively, the wrapper function mmtsne will wrap the matrix construction and multiple maps t-SNE model estimation into a single step.

no_maps

The number of maps (positive whole number) to be estimated.

no_dims

The number of dimensions per map. Typical values are 2 or 3.

max_iter

The number of iterations to run.

momentum

Constant scaling factor for update momentum in gradient descent algorithm.

final_momentum

Constant scaling factor for update momentum in gradient descent algorithm after the momentum switch point.

mom_switch_iter

The iteration at which momentum switches from momentum to final_momentum.

eps

A small positive value near zero.

Details

This code is an almost direct port of the original multiple maps t-SNE Matlab code by van der Maaten and Hinton (2012). mmtsne estimates a multidimensional array of N x no_dims x no_maps. Each map is an N x no_dims matrix of estimated t-SNE coordinates. When no_maps=1, multiple maps t-SNE reduces to standard t-SNE.

Value

A list that includes the following objects:

Y

An N x no_dims x no_maps array of predicted coordinates.

weights

An N x no_maps matrix of unscaled weights. A high weight on entry i, j indicates a greater contribution of point i on map j.

proportions

An N x no_maps matrix of scaled weights. A high weight on entry i, j indicates a greater contribution of point i on map j.

References

L.J.P. van der Maaten and G.E. Hinton. “Visualizing Non-Metric Similarities in Multiple Maps.” Machine Learning 87(1):33-55, 2012. PDF.

Examples

# Load the iris dataset
data("iris")

# Produce a symmetric joint probability matrix
prob_matrix <- p2sp(x2p(as.matrix(iris[,1:4])))

# Estimate a mmtsne model with 2 maps, 2 dimensions each
model <- mmtsneP(prob_matrix, no_maps=2, max_iter=100)

# Plot the results side-by-side for inspection
# Points scaled by map proportion weights plus constant factor
par(mfrow=c(1,2))
plot(model$Y[,,1], col=iris$Species, cex=model$proportions[,1] + 0.2)
plot(model$Y[,,2], col=iris$Species, cex=model$proportions[,2] + 0.2)
par(mfrow=c(1,1))