pg_optim: Proximal gradient optimization.

Description Usage Arguments Value

View source: R/opt_lasso.R

Description

Function for proximal-gradient optimization for non-differentiable functions We're trying to minimize f + g; where f is the differentiable part and g is the non-differentiable part.

Usage

1
pg_optim(x, f, df, pg, ts)

Arguments

x

Starting value.

f

Function to optimize.

df

Gradient of f.

pg

Proximal operator for g.

ts

Sequence of step sizes. use opt_ts() for stepsize generation

Value

The function returns the optimal solution


tathagatabasu/bootlasso documentation built on Aug. 9, 2019, 1:07 a.m.