nds_selection

Select best subset of points by non dominated sorting with
hypervolume contribution for tie breaking. Works on an arbitrary dimension
of size two or higher.
Non-dominated sorting is computed with <code>moocore::pareto_rank()</code> and
hypervolume contributions with <code>moocore::hv_contributions()</code>.
Boundary points, i.e., points that are best in at least one objective within their front,
always survive tie breaking.

internal

Features highly configurable search spaces via the 'paradox'
package and optimizes every user-defined objective function. The
package includes several optimization algorithms e.g. Random Search,
Iterated Racing, Bayesian Optimization (in 'mlr3mbo') and Hyperband
(in 'mlr3hyperband'). bbotk is the base package of 'mlr3tuning',
'mlr3fselect' and 'miesmuschel'.

Marc Becker

bbotk

Black-Box Optimization Toolkit

Jakob Richter

Michel Lang

Bernd Bischl

Martin Binder

Olaf Mersmann

nds_selection function

<dl><dt>points</dt>
<dd>(<code><a href="/link/matrix()?package=bbotk&version=1.13.0" data-mini-rdoc="bbotk::matrix()">matrix()</a></code>) 
Numeric matrix with each column corresponding to a point</dd>
<dt>n_select</dt>
<dd>(<code>integer(1L)</code>) 
Amount of points to select.</dd>
<dt>ref_point</dt>
<dd>(<code><a href="/link/numeric()?package=bbotk&version=1.13.0" data-mini-rdoc="bbotk::numeric()">numeric()</a></code>) 
Reference point for hypervolume.</dd>
<dt>minimize</dt>
<dd>('logical()') 
Should the ranking be based on minimization?
Can be specified for each dimension or for all.
Default is <code>TRUE</code> for each dimension.</dd></dl>

Arguments

Best points w.r.t. non dominated sorting with hypervolume contribution. — nds_selection

<dl>

<dt>points</dt>
<dd>(<code><a href='https://rdrr.io/r/base/matrix.html'>matrix()</a></code>) 
Numeric matrix with each column corresponding to a point</dd>


<dt>n_select</dt>
<dd>(<code>integer(1L)</code>) 
Amount of points to select.</dd>


<dt>ref_point</dt>
<dd>(<code><a href='https://rdrr.io/r/base/numeric.html'>numeric()</a></code>) 
Reference point for hypervolume.</dd>


<dt>minimize</dt>
<dd>('logical()') 
Should the ranking be based on minimization?
Can be specified for each dimension or for all.
Default is <code>TRUE</code> for each dimension.</dd>

</dl>

nds_selection: Best points w.r.t. non dominated sorting with hypervolume contribution.

Description

Usage

Value

Arguments