L1.ceiling {gelnet} | R Documentation |
Computes the smallest value of the LASSO coefficient L1 that leads to an all-zero weight vector for a given linear regression problem.
L1.ceiling(X, y, a = rep(1, nrow(X)), d = rep(1, ncol(X)), P = diag(ncol(X)), m = rep(0, ncol(X)), l2 = 1, balanced = FALSE)
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. |
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 |
l2 |
coefficient for the L2-norm penalty |
balanced |
boolean specifying whether the balanced model is being trained (binary classification only) (default: FALSE) |
The cyclic coordinate descent updates the model weight w_k using a soft threshold operator S( \cdot, λ_1 d_k ) that clips the value of the weight to zero, whenever the absolute value of the first argument falls below λ_1 d_k. From here, it is straightforward to compute the smallest value of λ_1, such that all weights are clipped to zero.
The largest meaningful value of the L1 parameter (i.e., the smallest value that yields a model with all zero weights)