Package dev.nm.solver.multivariate.unconstrained.c2.linesearch

Class FletcherLineSearch

java.lang.Object

dev.nm.solver.multivariate.unconstrained.c2.linesearch.FletcherLineSearch

All Implemented Interfaces:

LineSearch, Optimizer<C2OptimProblem,LineSearch.Solution>

public class FletcherLineSearch
extends Object
implements LineSearch

This is Fletcher's inexact line search method. An inexact line search is usually used in conjunction with a multi-dimensional optimization algorithm. It trades off the accuracy of the line search result with the amount of computation. A lot of optimization algorithms are shown to be quite tolerant to line search imprecision.

Nested Class Summary

Nested classes/interfaces inherited from interface dev.nm.solver.multivariate.unconstrained.c2.linesearch.LineSearch

LineSearch.Solution

Constructor Summary

Constructors

Constructor	Description
FletcherLineSearch()	Construct a line search minimizer using the Fletcher method with the default control parameters.
FletcherLineSearch(double rho, double sigma, double tau, double chi, double epsilon, int maxIterations)	Construct a line search minimizer using the Fletcher method.

Method Summary

Modifier and Type	Method	Description
LineSearch.Solution	solve(C2OptimProblem problem)	Solve an optimization problem, e.g., OptimProblem.

Methods inherited from class java.lang.Object

clone, equals, finalize, getClass, hashCode, notify, notifyAll, toString, wait, wait, wait

Constructor Detail

FletcherLineSearch

public FletcherLineSearch(double rho,
                          double sigma,
                          double tau,
                          double chi,
                          double epsilon,
                          int maxIterations)

Construct a line search minimizer using the Fletcher method.

Parameters:

rho - a precision parameter; smaller ρ, e.g., 0.1, gives better accuracy

sigma - a precision parameter; smaller σ, e.g., 0.1, gives better accuracy

tau - a control parameter to prevent the result from being too close to boundary

chi - a control parameter to prevent the result from being too close to boundary

epsilon - a precision parameter: when a number |x| ≤ ε, it is considered 0. This affects the precision of the result.

maxIterations - the maximum number of iterations. This affects the precision of the result.

FletcherLineSearch

public FletcherLineSearch()

Construct a line search minimizer using the Fletcher method with the default control parameters.

Method Detail

solve

public LineSearch.Solution solve(C2OptimProblem problem)

Description copied from interface: Optimizer

Solve an optimization problem, e.g., OptimProblem.

Specified by:

solve in interface Optimizer<C2OptimProblem,LineSearch.Solution>

Parameters:

problem - an optimization problem

Returns:

a solution to the optimization problem