gelnet {gelnet}

GELnet for linear regression, binary classification and one-class problems.

Description

Infers the problem type and learns the appropriate GELnet model via coordinate descent.

Usage

gelnet(X, y, l1, l2, nFeats = NULL, a = rep(1, n), d = rep(1, p),
  P = diag(p), m = rep(0, p), max.iter = 100, eps = 1e-05,
  w.init = rep(0, p), b.init = NULL, fix.bias = FALSE, silent = FALSE,
  balanced = FALSE, nonneg = FALSE)

Arguments

X

n-by-p matrix of n samples in p dimensions

y

n-by-1 vector of response values. Must be numeric vector for regression, factor with 2 levels for binary classification, or NULL for a one-class task.

l1

coefficient for the L1-norm penalty

l2

coefficient for the L2-norm penalty

nFeats

alternative parameterization that returns the desired number of non-zero weights. Takes precedence over l1 if not NULL (default: NULL)

a

n-by-1 vector of sample weights (regression only)

d

p-by-1 vector of feature weights

P

p-by-p feature association penalty matrix

m

p-by-1 vector of translation coefficients

max.iter

maximum number of iterations

eps

convergence precision

w.init

initial parameter estimate for the weights

b.init

initial parameter estimate for the bias term

fix.bias

set to TRUE to prevent the bias term from being updated (regression only) (default: FALSE)

silent

set to TRUE to suppress run-time output to stdout (default: FALSE)

balanced

boolean specifying whether the balanced model is being trained (binary classification only) (default: FALSE)

nonneg

set to TRUE to enforce non-negativity constraints on the weights (default: FALSE )

Details

The method determines the problem type from the labels argument y. If y is a numeric vector, then a regression model is trained by optimizing the following objective function:

\frac{1}{2n} \sum_i a_i (y_i - (w^T x_i + b))^2 + R(w)

If y is a factor with two levels, then the function returns a binary classification model, obtained by optimizing the following objective function:

-\frac{1}{n} \sum_i y_i s_i - \log( 1 + \exp(s_i) ) + R(w)

where

s_i = w^T x_i + b

Finally, if no labels are provided (y == NULL), then a one-class model is constructed using the following objective function:

-\frac{1}{n} \sum_i s_i - \log( 1 + \exp(s_i) ) + R(w)

where

s_i = w^T x_i

In all cases, the regularizer is defined by

R(w) = \lambda_1 \sum_j d_j |w_j| + \frac{\lambda_2}{2} (w-m)^T P (w-m)

The training itself is performed through cyclical coordinate descent, and the optimization is terminated after the desired tolerance is achieved or after a maximum number of iterations.

Value

A list with two elements:

w

p-by-1 vector of p model weights

b

scalar, bias term for the linear model (omitted for one-class models)

See Also

gelnet.lin.obj, gelnet.logreg.obj, gelnet.oneclass.obj