prox.sparse.func {LSVAR}

Proximal function with l1-norm penalty updating

Description

Proximal function with l1-norm

Usage

prox.sparse.func(w1, y, A, b, L, lambda, AtA, Atb)

Arguments

w1

previously updated model parameter

y

updated model parameter

A

design matrix

b

correspond vector, or matrix

L

learning rate

lambda

tuning parameter for sparse component

AtA

Gram matrix of design matrix A

Atb

inner product of design matrix A and correspond vector b

Value

Value of proximal function with l1-norm penalty