Implements adaptive lasso regularization for general purpose optimization problems. The penalty function is given by: $$p( x_j) = p( x_j) = \frac{1}{w_j}\lambda| x_j|$$ Adaptive lasso regularization will set parameters to zero if \(\lambda\) is large enough.
Usage
gpAdaptiveLasso(
par,
regularized,
weights = NULL,
fn,
gr = NULL,
lambdas = NULL,
nLambdas = NULL,
reverse = TRUE,
curve = 1,
...,
method = "glmnet",
control = lessSEM::controlGlmnet()
)Arguments
- par
labeled vector with starting values
- regularized
vector with names of parameters which are to be regularized.
- weights
labeled vector with adaptive lasso weights. NULL will use 1/abs(par)
- fn
R function which takes the parameters as input and returns the fit value (a single value)
- gr
R function which takes the parameters as input and returns the gradients of the objective function. If set to NULL, numDeriv will be used to approximate the gradients
- lambdas
numeric vector: values for the tuning parameter lambda
- nLambdas
alternative to lambda: If alpha = 1, lessSEM can automatically compute the first lambda value which sets all regularized parameters to zero. It will then generate nLambda values between 0 and the computed lambda.
- reverse
if set to TRUE and nLambdas is used, lessSEM will start with the largest lambda and gradually decrease lambda. Otherwise, lessSEM will start with the smallest lambda and gradually increase it.
- curve
Allows for unequally spaced lambda steps (e.g., .01,.02,.05,1,5,20). If curve is close to 1 all lambda values will be equally spaced, if curve is large lambda values will be more concentrated close to 0. See ?lessSEM::curveLambda for more information.
- ...
additional arguments passed to fn and gr
- method
which optimizer should be used? Currently implemented are ista and glmnet.
- control
used to control the optimizer. This element is generated with the controlIsta and controlGlmnet functions. See ?controlIsta and ?controlGlmnet for more details.
Details
The interface is similar to that of optim. Users have to supply a vector with starting values (important: This vector must have labels) and a fitting function. This fitting functions must take a labeled vector with parameter values as first argument. The remaining arguments are passed with the ... argument. This is similar to optim.
The gradient function gr is optional. If set to NULL, the numDeriv package will be used to approximate the gradients. Supplying a gradient function can result in considerable speed improvements.
Adaptive lasso regularization:
Zou, H. (2006). The adaptive lasso and its oracle properties. Journal of the American Statistical Association, 101(476), 1418–1429. https://doi.org/10.1198/016214506000000735
For more details on GLMNET, see:
Friedman, J., Hastie, T., & Tibshirani, R. (2010). Regularization Paths for Generalized Linear Models via Coordinate Descent. Journal of Statistical Software, 33(1), 1–20. https://doi.org/10.18637/jss.v033.i01
Yuan, G.-X., Chang, K.-W., Hsieh, C.-J., & Lin, C.-J. (2010). A Comparison of Optimization Methods and Software for Large-scale L1-regularized Linear Classification. Journal of Machine Learning Research, 11, 3183–3234.
Yuan, G.-X., Ho, C.-H., & Lin, C.-J. (2012). An improved GLMNET for l1-regularized logistic regression. The Journal of Machine Learning Research, 13, 1999–2030. https://doi.org/10.1145/2020408.2020421
For more details on ISTA, see:
Beck, A., & Teboulle, M. (2009). A Fast Iterative Shrinkage-Thresholding Algorithm for Linear Inverse Problems. SIAM Journal on Imaging Sciences, 2(1), 183–202. https://doi.org/10.1137/080716542
Gong, P., Zhang, C., Lu, Z., Huang, J., & Ye, J. (2013). A General Iterative Shrinkage and Thresholding Algorithm for Non-convex Regularized Optimization Problems. Proceedings of the 30th International Conference on Machine Learning, 28(2)(2), 37–45.
Parikh, N., & Boyd, S. (2013). Proximal Algorithms. Foundations and Trends in Optimization, 1(3), 123–231.
Examples
# This example shows how to use the optimizers
# for other objective functions. We will use
# a linear regression as an example. Note that
# this is not a useful application of the optimizers
# as there are specialized packages for linear regression
# (e.g., glmnet)
library(lessSEM)
set.seed(123)
# first, we simulate data for our
# linear regression.
N <- 100 # number of persons
p <- 10 # number of predictors
X <- matrix(rnorm(N*p), nrow = N, ncol = p) # design matrix
b <- c(rep(1,4),
rep(0,6)) # true regression weights
y <- X%*%matrix(b,ncol = 1) + rnorm(N,0,.2)
# First, we must construct a fiting function
# which returns a single value. We will use
# the residual sum squared as fitting function.
# Let's start setting up the fitting function:
fittingFunction <- function(par, y, X, N){
# par is the parameter vector
# y is the observed dependent variable
# X is the design matrix
# N is the sample size
pred <- X %*% matrix(par, ncol = 1) #be explicit here:
# we need par to be a column vector
sse <- sum((y - pred)^2)
# we scale with .5/N to get the same results as glmnet
return((.5/N)*sse)
}
# let's define the starting values:
b <- c(solve(t(X)%*%X)%*%t(X)%*%y) # we will use the lm estimates
names(b) <- paste0("b", 1:length(b))
# names of regularized parameters
regularized <- paste0("b",1:p)
# define the weight for each of the parameters
weights <- 1/abs(b)
# we will re-scale the weights for equivalence to glmnet.
# see ?glmnet for more details
weights <- length(b)*weights/sum(weights)
# optimize
adaptiveLassoPen <- gpAdaptiveLasso(
par = b,
regularized = regularized,
weights = weights,
fn = fittingFunction,
lambdas = seq(0,1,.01),
X = X,
y = y,
N = N
)
plot(adaptiveLassoPen)
# You can access the fit results as follows:
adaptiveLassoPen@fits
# Note that we won't compute any fit measures automatically, as
# we cannot be sure how the AIC, BIC, etc are defined for your objective function
# for comparison:
# library(glmnet)
# coef(glmnet(x = X,
# y = y,
# penalty.factor = weights,
# lambda = adaptiveLassoPen@fits$lambda[20],
# intercept = FALSE,
# standardize = FALSE))[,1]
# adaptiveLassoPen@parameters[20,]