rnso-package: Hybrid Algorithm for Non-Smooth Optimization (HANSO)

In rnso: An R Implementation of Hybrid Algorithm for Non-Smooth Optimization (HANSO)

Description

This package implements HANSO algorithm for optimization of smooth and nonsmooth functions. It uses both BFGS and gradient sampling methods. It also provides the users the option to use BFGS, gradient sampling and wolfe line search (both strong and weak).

Details

Package:	rnso
Type:	Package
Version:	1.0
Date:	2013-07-17
License:	MIT

The main functions of this package are:
hanso(): Hanso algorithm
bfgs(): Minimiza function using BFGS
gradsamp(): Gradient sampling algorithm for non smooth (and smooth) functions
linesch_sw(): Strong wolfe line search
linesch_ww(): Weak wolfe line search

Author(s)

Abhirup Mallik, University of Minnesota, Hans Werner Borchers

Maintainer: Abhirup Mallik <[e-mail]>

References

A.S. Lewis and M.L. Overton, Nonsmooth Optimization via BFGS, 2008.
J.V. Burke, A.S. Lewis and M.L. Overton, A Robust Gradient Sampling Algorithm for Nonsmooth, Nonconvex Optimization SIAM J. Optimization 15 (2005), pp. 751-779

Examples

fr <- function(x) {   ## Rosenbrock Banana function
  x1 <- x[1]
  x2 <- x[2]
  100 * (x2 - x1 * x1)^2 + (1 - x1)^2
}
grr <- function(x) { ## Gradient of 'fr'
  x1 <- x[1]
  x2 <- x[2]
  c(-400 * x1 * (x2 - x1 * x1) - 2 * (1 - x1),
    200 *      (x2 - x1 * x1))
}
#BFGS
(res=bfgs(fr,grr,nvar=2)) # using all the default parameters, using weak wolfe search
#HANSO
(res=hanso(fr,grr,nvar=2))
#Gradient sampling
x0 <- matrix(rnorm(12),2,6)
(tmp <- gradsamp(fr,grr,2,x0))
#Strong line search
(res=linesch_sw(fr,grr,c(-1.2,1),c(1,1)))
#Weak line search
linesch_ww(fr,grr,c(-1.2,1),c(1,1))

rnso documentation built on May 2, 2019, 6:12 p.m.