FSGFunL2: The Feature Selection and Grouping Function FSGFun with L2...

In ipapercodes/FGSPCA: Feature Grouping and Sparse Principal Component Analysis (FGSPCA)

Description

The FSGFun with L2 loss is a subproblem of the FGSPCA with L2 loss.

Usage

FSGFunL2(
  x,
  y,
  beta,
  tau_S,
  lambda1,
  lambda2,
  lambda3,
  v = 1,
  c = 1.02,
  iter_m_max = 100,
  iter_k_max = 50,
  condition_tol = 1e-05,
  nnConstraint = FALSE,
  sparseTruncated = TRUE,
  loss_return = FALSE
)

Arguments

x	the data matrix X_{n\times p}, where n is the number of observations, p is the number of features.
y	the response vector Y_{n\times 1} with length n
beta	β, the estimation of β
tau_S	τ, a global τ, which is assigned to τ_1=τ_2=τ .
lambda1	λ_1, the tuning parameter corresponding to p_1(\cdot)
lambda2	λ_2, the tuning parameter corresponding to p_2(\cdot)
lambda3	λ_3, the tuning parameter corresponding to p_3(\cdot)
v	the initial value of the Lagrange multiplier, with default v=1.0.
c	the acceleration constant to speed up the convergence procedure, default c=1.02.
iter_m_max	the maximum number of the outer iterations (i.e. the m-iteration), default iter_m_max=100
iter_k_max	the maximum number of the inner iterations (i.e. the k-iteration), default iter_k_max=50
condition_tol	the conditional tolerance of both the outer and inner iterations, default condition_tol=1e-5
nnConstraint	Boolean, indicating the non-negative constraint is true or false, default FALSE
sparseTruncated	Boolean, indicating whether to use the truncated L1 penalty or not for sparsity, default TRUE
loss_return	Boolean, indicating whether return the loss or not, default FALSE

Details

The parameters of x, y, beta, lambda1, lambda2, lambda3 are in inherit from ObjFunL2.

Value

the solution β or with the corresponding loss if loss_return=TRUE

ipapercodes/FGSPCA documentation built on Dec. 20, 2021, 7:58 p.m.