ts_change

<code>numeric</code> vector of last \(\gamma_n\) values, as selected by the
function passed to <code><a rd-options="=noisyCE2" href="/link/noisyCE2()?package=noisyCE2&version=1.1.0&to=%3DnoisyCE2" data-mini-rdoc="=noisyCE2::noisyCE2()">noisyCE2()</a></code> through the argument <code>stopwindow</code>.

relative changes smaller than <code>tol</code> stop the algorithm.

reltol

changes smaller than <code>tol</code> stop the algorithm.

Deterministic stopping rule based on the last change in the value of
\(\gamma_n\). Changes smaller than <code>tol</code>, or relative changes
smaller than <code>reltol</code> stop the algorithm. This criterion is suitable
only in case of deterministic objective functions.

Cross-Entropy optimisation of unconstrained deterministic and noisy
functions illustrated in Rubinstein and Kroese (2004, ISBN:
978-1-4419-1940-3) through a highly flexible and customisable function which
allows user to define custom variable domains, sampling distributions,
updating and smoothing rules, and stopping criteria. Several built-in
methods and settings make the package very easy-to-use under standard
optimisation problems.

Flavio Santi

noisyCE2

Cross-Entropy Optimisation of Noisy Functions

ts_change function

<code>numeric</code> vector of last \(\gamma_n\) values, as selected by the
function passed to <code><a rd-options='=noisyCE2' href='noisyCE2()'>noisyCE2()</a></code> through the argument <code>stopwindow</code>.

ts_change: Time series change stopping rule

Description

Usage

Arguments

Value

See Also